article
stringlengths 0
745k
| abstract
stringlengths 0
92.7k
|
|---|---|
the large hadron collider ( lhc ) is the last in the series of hadron colliders after the isr ( intersecting storage ring ) , sps ( super proton synchrotron ) , tevatron , hera and rhic ( relativistic heavy ion collider ) .
the machine elements are installed on average 100 m below the surface in the 26.7 km long accelerator tunnel that housed the large electron positron collider ( lep ) between 1989 and 2000 @xcite .
the ring consists of eight arcs and of eight long straight sections ( lsss ) .
the large particle physics experiments alice , atlas , cms and lhcb are installed at interaction points ( ips ) in the middle of four lsss , while the other lsss house the collimation ( or beam cleaning ) system , the radio - frequency ( rf ) system , the beam instrumentation and the beam dumping system .
the layout of the lhc is shown in fig .
[ fig : layout ] . a dipole field of 8.3 t is required to bend hadrons with a momentum of 7 tev/@xmath0 per unit charge in the tunnel ; this is 60% higher than in previous accelerators .
such a magnetic field strength is achieved with superconducting dipole magnets made of nbti . with a 2-in-1 magnet design the two rings fit inside the 3.8 m diameter lep tunnel
; see fig .
[ fig : magnet - cut ] .
both rings are accommodated in a single cryostat and the distance between the two vacuum chambers is only 19 cm .
the two proton or ion beams circulate in opposite directions in two independent vacuum chambers .
each dipole magnet is 14.3 m long ; the associated cryostat is 15 m long . besides the 1232 dipole magnets that constitute around 85% of each arc
, the magnet lattice also includes quadrupole magnets that focus the beam , sextupole magnets to correct chromatic effects and octupoles to stabilize the beam .
a total of 8000 superconducting magnets are used to control the two beams .
eight continuous cryostats with a length of 2.7 km each cool the superconducting magnets to their operating temperature of 1.9 k. after cool down the lhc cryostats contain 130 tons of liquid helium and around 37000 tons of material are cooled to that temperature .
the magnets and the cooling system based on superfluid helium form by far the longest refrigerators on earth .
the cryogenic system has to be extremely reliable ; in 2012 the system achieved an overall uptime of 95% . around 1600 power converters
provide current to the magnets ; for the main circuits the peak currents reach 13 ka . the magnetic energy stored in each arc cryostat is around 1 gj .
this energy has to be safely extracted in case one of the magnets quenches , i.e. performs a transition from the superconducting to the normal - conducting state @xcite .
large dump resistors that are capable of absorbing the energy are automatically switched into the main circuits in case of a quench .
the performance of a collider is characterized by its luminosity @xmath1 .
the event rate of a physical process with cross - section @xmath2 ( with unit of an area [ m@xmath3 ) is given by @xmath4 .
the luminosity may be expressed as @xmath5 where @xmath6 is the revolution frequency ( 11.24 khz for the lhc ) , @xmath7 is the number of bunches , @xmath8 is the number of particles per bunch and @xmath9 and @xmath10 are the horizontal and vertical beam sizes at the collision point .
the highest luminosity is achieved with the smallest possible beam cross - section , a large number of bunches and a high bunch population . up to 2808 bunches
can be filled and accelerated in each lhc beam ; the minimum distance between bunches is 25 ns or 7.5 m. each proton bunch consists of up to @xmath11 protons .
the bunches have a typical length of 7 to 10 cm . at the interaction
point the transverse rms sizes of the colliding beams are around 20 @xmath12 m @xcite .
figure [ fig : lumicolliders ] presents the evolution of hadron collider luminosity over time .
the lhc pushes the energy frontier by a factor of seven and the luminosity frontier by a factor of 25 ; the luminosity gain is mainly obtained with very high beam intensities .
the lhc dipole magnets were produced by three industrial firms and the last dipole magnet was delivered to cern in november 2006 .
each magnet was trained on cern test benches to a magnetic field of 8.7 t , approximately 5% above the design target .
a few training quenches were typically required to reach the nominal field of 8.3 t. training quenches are due to the release of extremely small amounts of frictional energy ( 10100 nj ) due to coil movements when the magnetic field is increased . in june 2007 the first arc ( sector ) of the lhc was cooled down and ready for commissioning and in april 2008 the last dipole magnet was lowered into the lhc tunnel .
one of the essential components of the commissioning phase was the testing of the lhc superconducting magnets and the associated powering and protection equipment .
in early 2008 it became apparent that the lhc dipole magnets had to be re - trained to their nominal field ; the first magnet quenches appeared at fields corresponding to beam energies of around 5.5 tev .
a training campaign on one arc revealed that the number of required re - training quenches increased rapidly with the magnetic field .
the estimated number of quenches required to reach 6.5 tev is around 140 , confirmed during the re - commissioning in 2015 , while for 7 tev the expected number can be as high as 1000 .
since such a training campaign would have required a long time , it was decided to lower the energy for the commissioning and first operation phase to 5 tev @xcite . on 10
september 2008 beams were circulating for the first time in both lhc rings .
the startup was however brought to an abrupt halt on 19 september 2008 when a defective high - current soldering between two magnets triggered an electric arc that released around 600 mj of stored magnetic energy @xcite .
the accelerator was damaged over a distance of around 700 m ; 53 magnets had to be replaced or repaired .
the beam vacuum was polluted with dust and soot over 2 km .
the repair , improvement and re - commissioning of the lhc lasted until november 2009 .
more details on the incident that did not involve beams are presented in the appendix . during the repair a systematic soldering problem affecting roughly 15% of all the high - current ( 13 ka )
cable joints was discovered . since this problem could not be solved immediately , the beam energy had to be limited to 3.5 tev until a complete repair campaign could be performed .
the lhc therefore operated in 2010 and 2011 at this energy , before the energy was increased to 4 tev in 2012 and 2013 because no magnet was quenched at 3.5 tev during beam operation , thus lowering the risk of problems with the cable joints .
the energy will be pushed to 6.5 tev and above in 2015 after the repair campaign of 2013 and 2014 @xcite .
at 7 tev each nominal lhc beam stores an energy of 360 mj .
this corresponds to the energy content of 80 kg of explosives and is a hundred times higher than previously achieved accelerator records .
the energy stored in the beams of various accelerators is shown in fig . [
fig : storede ] . machine protection systems ( mpss ) with unprecedented safety levels are therefore required to operate the lhc @xcite .
this is achieved with a combination of active protection by equipment and beam parameter monitoring , as well as with passive protection by a large number of collimators as indicated in fig .
[ fig : mps - function ] .
the function of the lhc mps is to protect the lhc accelerator , its injection and extraction transfer lines , as well as its experiments @xcite against a large spectrum of failures @xcite .
failures may be detected by monitoring equipment or beam parameters , for example power converter currents and state or beam loss rates . in parallel their impact
is mitigated by passive absorbers in the form of collimators and absorbers . in the case where abnormal conditions are detected ,
interlock signals are sent to the the beam interlock system ( bis ) @xcite that will trigger a beam dump or stop injection of the beams .
a summary of the time - scale of failures and of the reaction time of mps components is shown in fig .
[ fig : mps - timescale ] . the lhc bis interfaces the sources of interlocks ( clients ) and the beam dumping and injection kicker systems .
there are currently 189 inputs from client systems .
the list of connected systems includes : * the powering interlocks for normal - conducting and superconducting circuits ; * the fast current change detection of electrical circuits with fast failure time constants ; * the electrical circuits of the main experiment magnets ; * the interlock signals from the experiments ( excessive rates ) ; * approximately 4000 beam loss monitors ( blms ) @xcite ; * the beam position monitors ( bpms ) ; * the beam screens ; * the collimators and absorbers ( positions and temperatures ) ; * the movable experimental detectors ; * the rf system ; * the beam dumping system ( state and trigger unit ) ; * the injection and aperture kicker systems ; * the personnel access system and its associated beam stoppers ; * the operator inhibit buttons ; * the vacuum valves ; * the programmable beam dump .
beam interlock controller ( bic ) modules are installed in each lhc access point and provide the interface between the client systems and the bis ; see fig .
[ fig : bisloop ] .
the bic units are connected to each other and to the beam dumping system though permit loops .
there are two permit loops per beam ; the signals in the loops propagate clockwise in one loop and counter - clockwise in the other loop .
this arrangement ensures always the fastest possible transmission of a dump request to the dumping system . at the lhc
the dump delay can reach around three turns or 270 @xmath12s ; this delay includes propagation of an interlock on the permit loops , synchronization of the beam dump with the beam abort gap ( see below ) and one turn to extract the entire beam .
the permit loops are connected directly to the lhc injection interlock system , which is made from the same building blocks ( bic modules ) .
the lhc injection interlocks act on the lhc injection kicker and on the sps extraction interlock system .
this ensures that no beam may be extracted from the sps injector when the beam permit loops are not armed @xcite .
the superconducting magnets in the lhc lose their superconducting state ( quench ) when they are heated with about 10 mj per cubic centimetre .
this is to be compared with the stored energy of many 100 mj . to prevent unavoidable particle losses from the beam hitting the vacuum chamber within the magnets , possibly leading to sufficiently large energy deposition to trigger a quench @xcite
, a collimation system @xcite must intercept the beam losses with very high efficiency .
contrary to previous hadron colliders that used collimators only for experimental background conditions , the lhc can not operate without its collimation system .
approximately 100 lhc collimators are installed in two long straight sections as shown in fig .
[ fig : layout ] . to be able to absorb the energy of the 7 tev hadrons , the lhc requires a multistage collimation system that intercepts the particles in a four - step process as outlined in fig .
[ fig : cleaning ] , covering the entire six - dimensional phase space , catching particles with large transverse oscillation amplitudes and energy errors @xcite .
each collimator is composed of two parallel , typically 1.2 m long material blocks ( jaws ) that define a variable gap through which the beams circulate .
the smallest gap widths are roughly 2 mm . depending on their location in the ring and on their role ,
the collimator jaws are made of fibre - reinforced carbon , strongly absorbing copper and tungsten blocks .
the system worked perfectly so far , also thanks to excellent beam stability and machine reproducibility . around 99.99% of the protons that were lost from the beam
were intercepted . a single set - up of the collimation system
was required per year . from the machine protection perspective
, the lhc collimator must fulfil a dual role , halo collimation ( beam cleaning ) and passive protection of the accelerator component .
the lhc beam dumping system ( lbds ) is a complex system composed of 15 fast kicker magnets that deflect the beam to the outside of the ring , 15 normal - conducting septum magnets that deflect the beam vertically out of the plane of the ring , 10 dilution kicker magnets that spread the beam over the beam dump surface and finally the beam dump block @xcite .
the layout of the lbds is shown in fig .
[ fig : lbds - layout ] .
the 10 m long carbon dump block is the only lhc element capable of absorbing the nominal beam .
the beam is swept over the dump surface to lower the power density . without the sweep
the beam could drill a hole with a depth of a few metres into the dump block through hydro - dynamic tunnelling @xcite .
a 3 @xmath12s long particle - free gap in the beam ( beam abort gap ) provides a time window for the lbds dump kickers to raise to their nominal field .
the dump kickers must be accurately synchronized to the beam abort gap to avoid spreading beam across the aperture during the kicker rise time .
possible failure modes are : * the abort gap fills with beams ( rf fault , debunching , injection error ) ; * the kicker synchronization fails or a kicker fires spontaneously ( not synchronized ) so - called _ asynchronous beam dumps_. the asynchronous dump is the _ ultimate unavoidable failure _ : the lhc must be protected from this failure _ passively _ by absorbers .
two large absorbers in front of the extraction septum and in front of the first superconducting magnet downstream of the dump kickers and septa protect the lhc against damage from asynchronous dumps and from residual beam in the abort gap .
dump kicker powering , synchronization and triggering are designed to exclude out - of - sync triggers with high reliability .
the spontaneous trigger is estimated to occur roughly once per year .
so far none has been observed during high - intensity operation , but the system operated at reduced high voltage ( 4 tev instead of 7 tev ) , which also lowers the probability of spontaneous kicker firing . a direct link between the lbds and the injection system ensures that the lhc injection kicker magnets can not be triggered with a timing that would inject beam into the abort gap .
the beam population in the abort gap is monitored using synchrotron light .
when the particle density in the abort gap exceeds a pre - defined threshold , the transverse feedback system of the lhc is used to excite those particles until they hit the collimators .
already during the design phase of the lhc mps , the need for masking interlocks was recognized @xcite .
some flexibility is always required for low - intensity commissioning and setting up phases where the risk of damage is also much lower .
to avoid masking interlocks by raising thresholds and opening tolerance windows for many parameters ( with the risk of errors during the reversal ) , the concept of masking interlocks when _ the beam is safe _ was introduced at an early stage .
such a safe beam should not be able to damage accelerator components .
the corresponding beam intensity limit depends on the beam energy and on the beam emittance .
it also depends on the material that is considered for damage . at the lhc
, copper was selected as reference material .
given the large difference between the energy required to quench a superconducting magnet and the energy required to damage the same magnet , it is clear that even a safe beam will be able to quench one or more magnets .
in the early design phase of the lhc machine protection system , around the years 20002004 , not much was known about equipment damage by high - energy and high - intensity beams . in order to benchmark damage by a high - energy lhc - type beam
, a dedicated experiment was set up in a sps transfer line @xcite . in this
controlled experiment beams of 450 gev protons with lhc bunch structure ( 25 ns spacing ) and nominal lhc beam emittances were directed into a special target composed of a sandwich of materials ( copper , stainless steel and zinc ) . from this experiment
the damage limit for copper was established at @xmath13 protons .
it should be noted that the damage limit for stainless steel was more than a factor of four higher .
the observations were in good agreement with rather simple estimates for the damage limit based on shower simulations to define the deposited energy and heating of the materials up to the melting point .
fluka simulations @xcite were used to extrapolate the experimental results from 450 gev to the lhc operating energy of 7 tev .
the simulations predicted the following scaling law for the safe beam intensity @xmath14 with beam energy @xmath15 : * larger energy deposition implies a scaling @xmath16 ; * shrinking of the transverse emittance ( beam area ) implies a scaling @xmath16 ; * increase of the shower length provides some dilution @xmath17 .
the following effective scaling law with energy ( at fixed normalized emittance ) was obtained : @xmath18 this equation is implemented inside the lhc safe machine system ( smp ) and is indicated as the _ normal _ set - up beam flag ( sbf ) in fig .
[ fig : sbf - energy ] . at injection
the intensity limit was set initially to @xmath19 charges , but was later lowered to @xmath20 because the beam emittances were roughly a factor of two smaller than expected in the lhc design .
the smp system is connected to reliable beam current transformers ( bcts ) and energy sources ( based on the dipole fields with four - fold redundancy ) .
it generates the sbf that is distributed over the lhc machine timing system to the bis .
the sbf states are : * true = set - up beam : maskable bis input signals can be masked ; * false = unsafe beam : no bis input signals may be masked .
the beam interlock system is configured to allow masking certain classes of interlocks ( maskable ) when the sbf is true . during initial operation
it was realized that the definition of the sbf was too restrictive for beam commissioning that required higher bunch charges than expected ( nominal bunches with @xmath21 protons ) due to systematic effects in the lhc beam instrumentation , mainly the beam position monitoring system . as a consequence ,
relaxed sbf equations were introduced to be able to mask certain interlocks with 23 nominal bunches at top energy as shown in fig .
[ fig : sbf - energy ] . during commissioning periods
a mps expert can relax the equation of the sbf within the smp system for certain tests .
charges ) as a function of energy for the set - up beam flag .
the _ normal
_ sbf corresponds to the scaling of eq .
( [ eq : sbf - scaling ] ) .
the _ beam set - up _ equation is a relaxed version used for beam commissioning and mps set - up . for ion beams
the limits are scaled by the ratio @xmath22 of the charge per nucleon . ]
the mps activities of the lhc were organized since the year 2000 inside a machine protection working group ( later renamed to machine protection panel , mpp ) .
this group steered the design and followed up on implementation , issues and performance of the lhc mps .
all groups involved in mp activities were represented in the mpp .
an executive mp body , the restricted mpp ( rmpp ) , was created when beam commissioning started .
each core mp system has one representative within the rmpp .
the rmpp takes decisions related to mps ( example : blm threshold changes ) and steers the intensity ramp up of the machine .
its recommendations are submitted to the cern management ; in almost all cases the recommendations were accepted . before the machine startup
, procedures were developed for the commissioning of the machine protection subsystems .
the procedures contain test descriptions and frequency of tests ( after stop or intervention ) .
the procedures were then translated into a series of individual tests to be performed on the machine with and without beam , if required also at different intensity steps .
machine protection tests are currently documented and tracked on a web page .
one mps expert of the commissioning team checks that all tests required for a given commissioning phase have been performed by the experts .
it is in general the system expert that executes the tests for his system and not an independent person .
this simple mechanism for tracking the commissioning will be improved in the future .
the new concept with test tracking and electronic expert signatures is already in place for the commissioning of the lhc electrical circuits and magnets .
the _ acctesting _ framework @xcite is based on pre - defined and agreed test sequences . tests that are ready for execution can be scheduled for execution ; test sequences are blocked until tests are analysed and signed by experts .
the results are tracked and stored ; no test step can be forgotten . unfortunately due to lack of resources
the test tracking was not yet implemented for the lhc mps , but it is foreseen in the near future .
passive protection of the lhc aperture with collimators and absorbers is a key ingredient for operating the lhc safely at high intensity .
all failures affecting the machine on a global scale ( orbit , optics , emittance , perturbations ) must be intercepted by a protection device .
the lhc machine setting up involves : * a well - corrected orbit ( rms below 0.5 mm ) ; * a well - corrected optics ( betatron function beating below 20% at injection and below 10% with colliding beams ) ; * a good knowledge of the aperture bottlenecks ( after orbit and optics correction ) .
the machine aperture is measured after orbit and optics corrections have been finalized .
the global aperture is measured at injection energy ( typical aperture limit is at @xmath23 @xmath2 , where @xmath2 is the beam size with a nominal normalized emittance of 3.5 @xmath12 m ) .
local apertures are measured at injection ( injection region , beam dump section and around the four experiments ) and at top energy with fully squeezed beams ( only the four experiments ) . at top energy
the aperture depends on the optics at the different collision points .
the aperture limit is generally located in the low - beta quadrupoles next to the collision points . during machine set - up
, all collimators and absorbers are aligned around the closed orbit with appropriate retractions . for good performance
the orbit must be reproducible at the level of 50 @xmath12 m .
the machine set - up ( orbit , optics , aperture and protection devices ) is then validated by a campaign of loss maps and simulated asynchronous beam dump tests . * during a _ loss map _ the transverse beam emittance is blown up until losses are observed on the collimators and absorbers ; an example is shown in fig .
[ fig : lossmap ] .
the loss distribution provides a validation of the collimator alignment and hierarchy @xcite .
initially the emittance was blown up by crossing the third - order resonance in the horizontal or vertical plane .
this technique was however not always very reproducible ; sometimes the losses were too small , sometimes they were massive and a large fraction of the beam was lost . from 2012
onwards emittance blow up was obtained by noise excitation using the transverse feedback system .
this provided fine control over the losses and blow up could be applied to individual bunches @xcite .
* for a simulated _ asynchronous test dump _ a low - intensity beam ( typically 13 bunches ) is de - bunched by switching of the rf system .
after a few minutes the particles drift into the abort gap , the density of particles in the gap being monitored by a dedicated device based on synchrotron light . when the population in the abort gap is sufficient , a dump is triggered
the beam present in the region of the abort gap mimics the effect of an asynchronous dump .
the loss distribution along the ring provides a validation of the dump protection alignment .
the entire lhc cycle , from injection to collisions , is always set up with low - intensity beams .
the maximum intensity is three bunches with nominal intensity ( @xmath24 protons ) .
the set - up is made with less than 1 permill of the nominal intensity , which represents a challenge for the beam instrumentation , since there should not be a significant bias of measurements between the set - up and the nominal full - intensity beam ( for example for the beam position monitoring ) .
the intensity increase is steered through the restricted machine protection panel ( mppr ) , which defines the intensity steps and the requirements to proceed with the intensity increase .
the plan for the first learning year in 2010 foresaw three phases : * low - intensity beams for commissioning and early experience .
this phase was followed by an internal review of the mps performance ; * an intensity ramp up to a stored energy of 12 mj followed by 4 weeks of operation at that stored energy .
such a stored energy corresponded to state of the art in 2010 .
this phase was followed by an external review of the mps performance ; * increase of the stored energy into the regime above 10 mj .
figure [ fig : storede - run1 ] shows the evolution of the peak stored energy for one lhc beam between 2010 and 2012 .
the slow ramp up of 2010 is clearly visible . with the experience that was accumulated in 2010 ,
the stored energy was ramped up above 100 mj in 2011 over 11 intensity steps , up to a maximum of 1300 bunches .
the 2011 intensity ramp up took around 9 effective weeks ; the rate was dictated by operational ( and not mp ) issues as soon as around 600 bunches were stored @xcite .
losses and the need to adjust of blm thresholds , vacuum element heating by the beam , beam stability , etc slowed down the pace . finally , the 2012 intensity ramp up took just 2 weeks with seven intensity steps , as can be seen in fig .
[ fig : rampup-11 - 12 ] .
injection into a synchrotron that has no circulating beam has the same mp issues as a linac @xcite .
injection of an intense beam can represent a serious risk or require very important monitoring efforts ( all power converters etc ) to ensure that the beam is not lost directly on the aperture during the first turn , before the mps is able to react . to overcome this issue , the concept of _ witness _ beam ( or _ beam presence _ )
was introduced for the lhc @xcite .
the principle is that only a probe bunch ( typically @xmath25 protons , max @xmath21 protons ) may be injected into an _ empty _ ring .
such a low intensity can not provoke damage to components .
a high - intensity injection requires a minimum beam intensity to be circulating ; this is the best guarantee that the conditions are reasonable to avoid failures happening on the first turn , before the mps is able to react . at the lhc
the concept is based on a highly reliable and redundant intensity measurement .
a flag indicating the beam presence ( true / false ) is transmitted to the extraction interlock system of the sps injector where it is combined with a flag indicating that the sps beam is has a probe intensity ( max @xmath21 protons ) . despite storing up to 140 mj at 4 tev ,
not a single superconducting magnet was quenched at the lhc with circulating beam , despite quench thresholds of only a few tens of mj .
on the other hand , many magnets were quenched at injection , mainly due to ( expected ) injection kicker failures ( seven events in 2012 ) . the beam ( roughly 2 mj ) is safely absorbed in injection dump blocks , but the shower leakage can quench magnets over a distance of around 1 km . at the lhc characteristic beam losses are observed in the various phases of the machine cycle .
those losses are part of regular operation and must be tolerated , even if one tries to minimize them .
the normal losses are : * injection losses ( tails of the injected beams , injection oscillations , de - bunched beam ) ; * start of ramp losses ( uncaptured beam loss ) ; * scraping on collimators ( gap changes , orbit and tune shifts ) ; * losses from the beam halo when beams start to collide ( beam beam effect ) ; * losses due to the beam burn - off that are proportional to luminosity and performance .
the different types of losses are shown in fig .
[ fig : loss - cycle ] for a typical lhc cycle .
the importance of the primary collimator opening appears clearly if the total beam intensity transmission presented in fig .
[ fig : transmission ] is compared between 2011 and 2012 . in 2012
the collimators were set closer to the beam in order to protect a smaller aperture , which allowed smaller betatron functions to be reached at the ip and therefore smaller beam sizes at the collision points , yielding a 60% higher luminosity @xcite .
the gap width has a strong impact on beam transmission and losses in the cycle . * in 2011 the intensity losses during the ramp and squeeze phases are so small that they can not be measured with the current transformers .
the losses are completely dominated by collisions .
the primary collimator gaps correspond to an opening of @xmath26 @xmath2 , where @xmath2 is the real beam size . * in 2012 , on the other hand , there are already significant losses during the ramp when the collimator gaps are closed to their high - energy setting .
there are noticeable losses in the optics squeeze due to an increased sensitivity to even small orbit jitter at the level of 50 @xmath12 m .
the primary collimator gaps correspond to an opening of @xmath27 @xmath2 , where @xmath2 is the real beam size . the regular loss distribution during collisions is shown in fig .
[ fig : loss - sb ] .
the beam loss rates near the experiments are almost as high as at the collimators .
this is due to the fact that around the experiments the loss monitors record collision debris , the physics processes at very small angles that are not covered by the experiments .
their signals are proportional to the collider luminosity .
the thresholds of the lhc loss monitors are set to prevent quenches for monitors installed on superconducting elements and to prevent damage for room - temperature elements ( for example collimators ) @xcite .
in both cases some safety margin is desired .
the initial thresholds were set before lhc operation started based on rather coarse quench level estimates coupled to geant , fluka and mars simulations @xcite . during the first years of operation the thresholds were progressively adapted ( many were increased ) based on experience . initially the thresholds on collimators were set to limit the average power loss significantly below the peak design power of 500 kw .
the thresholds were only increased to match the nominal power loss once the operation of the lhc was well controlled and understood .
quench tests with wire scanners ( nice point - like particle source ) , orbit bumps and short and high losses in the collimation area were used to determine more accurately the quench limits @xcite .
very fast beam loss events ( time - scale of a millisecond and below ; see fig .
[ fig : ufo - time ] ) mainly in superconducting regions have been the _ surprise of lhc operation_. those fast losses were nicknamed ufos ( unidentified falling objects ) when it became clear that the ufo - type beam losses were due to small objects falling into the beam , the subsequent interaction of the beam particles leading to the showers and the beam losses @xcite .
the blm signals are consistent with small ( tens of @xmath12 m diameter ) dust particles ` entering ' the beam .
the vast majority of ufo events lead to losses below dump threshold , but around 20 beam dumps were triggered by such ufo - type events every year between 2010 and 2012 .
ufo events localized within the injection kickers could be traced to aluminium oxide dust present in large quantities on the inner surface of the ceramic vacuum chamber of those elements .
a cleaning campaign of the kickers was made during the long shutdown in 20132014 .
ufos occur mainly with high - intensity beams , but there is conditioning with beam ; the event rate drops from 10 to 2 per hour over a year as shown in fig .
[ fig : ufo - rate ] . to improve the sensitivity of the lhc arc blms to ufo events ,
two out of six blms installed around the arc quadrupoles were re - located to the dipoles during the shutdown @xcite .
the monitoring and protection ( quench prevention ) capabilities of the blms are significantly improved with this re - location for the coming 6.5 tev run . while the beam losses will increase due to the higher beam energy ,
the quench thresholds of the magnets will come down by a factor of 34 , making the situation at 6.5 tev more critical .
the three main requirements for a modern mps are : * protect the machine : the highest priority is obviously to avoid damage to the accelerator ; * protect the beam : complex protection systems reduce the availability of the accelerator ; the number of ` false ' interlocks stopping operation must be minimized .
this implies sometimes a trade - off between protection and operation ; * provide the evidence : clear ( post - mortem ) diagnostics must be provided when the protection systems stop operation or when something goes wrong ( failure , damage , but also ` near misses ' ) . once the mps components have been commissioned , it is essential that the protection functionality is maintained during operation .
automated checks of the mps components as pre- or post - flight checks can ensure that the mps functionality is not degraded . for colliders with long cycle times
there are two types of checks that fit well into the cycle , namely pre - flight checks before injection and post - flight checks on data collected during a fill or during the beam dump ( post - mortem data ) .
such tests can come in multiple forms , for example the verification of mps - related settings such as interlock thresholds , configuration checks of the beam interlock systems , automated analysis of the faults and mps reaction chain and automated analysis of the dump system action . at the lhc
the blm system integrity checks must be performed at least once per 24 h period ( or after the following beam dump if that interval is exceeded ) @xcite .
the integrity checks are performed by a high - voltage modulation that is analysed to detect ` dead ' channels . at the lhc
the mps is so critical that for every beam dump post - operation checks ( pocs ) are performed on the beam dump system post - mortem data ( equipment and beam signals ) .
data collection and analysis are triggered automatically after each dump .
the analysis covers internal beam dump system signals and external beam information like dumped beam intensities , losses and beam positions in and around the dump channel .
the analysis ensures that all signals are correct and that there is no loss of redundancy ; the beam dumping system can then be considered ` as good as new ' .
machine operation is stopped if the beam dumping system pocs fail ; an expert verification is required to re - start beam operation .
this concept was so successful that it was extended to lhc injection in the form of automated checks of each injection quality . already during the design of the lhc mps ,
post - mortem ( pm ) diagnostics was identified as a key component to understand the root causes of beam dumps @xcite .
all key lhc systems implement circular pm data buffers that are frozen and read out when a beam dump is triggered .
the pm trigger is transmitted over the lhc machine timing system .
the pm system also inserts an automatic entry in the lhc electronic logbook .
sampling frequencies of the data in the pm buffers range from ms or turn level to tens of milliseconds and are adapted to each system .
synchronization is critical to make sense of the data and define the event sequence ; therefore time stamping is made at the source of the data .
the lhc post - mortem data volume is currently around 200 mbytes . since the analysis of the large volume of lhc pm data can be tedious , automatic analysis tools
have been implemented to support operators and mps experts in their work .
the lhc pm server and graphical user interfaces are based on the java language with standard interfaces to extract raw data and provide analysed data @xcite .
this allows many persons to contribute to the analysis modules . after a beam dump ,
injection into the lhc is blocked until the pm data is collected , pre - analysed ( automatic ) and signed by the operation crews .
if an automated analysis identifies a critical problem , injection can only be released by a mps expert . to detect any possible long - term trend mps experts re - analyse all pm events collected above injection energy within a few days .
this also provides a view that is independent of the operation teams . depending on its size , complexity and energy range an accelerator has a large volume of mps settings in the form of blm thresholds , current references and tolerances , etc .
the management of settings associated to the mps poses special problems , since access to the settings must be limited to authorized experts , yet it must be possible to inspect and download the settings from a central database to the front - end systems of the mps . for the lhc access restrictions
have been put in place based on _ role - based access control _ ( rbac ) @xcite and on digital signatures . to protect mps settings the concept of _ management of critical settings _ ( mcs ) @xcite was developed , a settings protection system that is fully embedded in the controls middleware and settings management . a setting that is defined as _
critical _ has an associated digital signature .
only a user that has the appropriate rbac role ( mps expert , blm expert , etc ) is able to generate the digital signature .
the digital signature is generated at the moment when a setting is changed by an authenticated expert .
the setting and the associated digital signature are transmitted together to the front - end computer : a critical setting is only accepted with a valid digital signature , see fig .
[ fig : mcs ] .
the lhc mps has inputs from many systems that operate independently and there is only very limited information exchange between the various systems . to implement interlocks on a global machine scale with correlation of data between many lhc systems , a software interlock system ( sis )
was developed @xcite .
this system is able to collect information from any control device of the lhc and its injectors
. the lhc sis can dump one or both beams and inhibit lhc injection .
it may perform global scale analysis among systems , correlate injector and lhc ring data for injection protection , etc .
new interlock tests can be implemented rapidly to protect against unexpected issues .
the lhc sis is by design rather slow ( reaction time of the level of seconds ) but it is able to detect anomalies that could lead to problems in the future or prevent unnecessary beam dumps at injection .
the sis is based on a java core server , with a large data buffer for currently up to around 2500 devices and settings . for each device data
a timeout and no - data policy is defined . over 5000 tests
are defined with the lhc sis ; tests can be defined as simple value comparisons or complex java logic .
peak performance in terms of luminosity is not the most important lhc parameter ; the integrated performance is much more critical . to optimize the integrated time with collisions it is important to maximize the time spent colliding the beams and to minimize the time to re - establish collisions at the end of a fill ( turn - around time ) .
the fraction of time that the lhc spends colliding beams stably for the experiments amounted to 3035% of the total scheduled operation time @xcite .
the average duration of collisions was only 6 h while the optimum duration would be in the range of 8 to 12 h. the reason why the fills are so short is because during high - intensity operation only one out of three fills is dumped by the operation crews ; see fig .
[ fig : dump2012 ] .
the other two - thirds of the fills are dumped by the lhc mps @xcite .
roughly 14% of the beam dumps are due to the failures of mps subsystems ( ` false ' dumps ) , with the following distribution of the mps subsystems : * quench protection system ( radiation to electronics ) : 65% ; * blm system : 13% ; * beam dumping system : 12% ; * software interlock system : 5% ; * powering interlock system : 2.5% ; * beam interlock system : 1.5% .
a reliability working group predicted the rate of false dumps and the safety of the lhc mps for 7 tev operation before the lhc was switched on .
this can now be compared with observations , but it must be observed that so far lhc operated at 4 tev while the predictions were made for 7 tev , which is not completely equivalent for certain systems .
the observations are roughly in line with predictions , but some failures do not match completely ; in particular , radiation to electronics was not included in the initial predictions @xcite .
machine experiments can be very exciting , but also risky periods for an accelerator . frequently the machine is operated at some ` distance ' from standard conditions .
for example collimator settings , orbit and optics may be changed . at the lhc
every experiment is categorized according to the foreseen changes to the machine and to the beam intensity .
experiments using intensities above the sbf limit must be prepared with a detailed written description of the changes to machines and the test procedures . in many cases
the analysis of the document helped improve the efficiency of the experiment by spotting ` impossible ' things .
this encourages experimenters to think about options with smaller mp footprint , for example lower intensity .
the lhc mps has been very successful in protecting the lhc with over 100 mj of stored beam energy while the lhc was operated at 4 tev .
no component was damaged by a failure leading to beam loss ; the mps therefore fulfilled its job .
as expected , operation of the lhc was significantly constrained by mp due to the high stored energy and very low quench levels . nevertheless , operation was rather smooth and the intensity ramp up from well below 1 mj during the commissioning phase to over 100 mj stored energy took only 2 weeks in 2012 . in 2015
the beam energy will be increased to 6.5 tev : the stored energy of the beams will increase by a factor of 2 to 3 , while the quench levels of the magnet drop by a factor of 3 to 5 .
this will be a new challenge for lhc operation , while at the same time the focus is shifting more and more towards high(er ) machine availability .
99 l. evans , _ the large hadron collider : a marvel of technology _
( epfl press , lausanne , 2009 ) .
o. brning et al .
( eds ) , lhc design report , cern-2004 - 003 ( 2004 ) .
http://dx.doi.org/10.5170/cern-2004-003-v-1 .
h. pfeffer , protection of hardware : powering systems ( pc , nc and sc magnets ) , these proceedings .
j. wenninger et al . , operation and configuration of the lhc in run 1 , cern - acc - note-2013 - 0041 ( 2013 ) .
l. rossi , the large hadron collider of cern and the roadmap toward higher performance , cern - acc-2014 - 0225 .
r. schmidt et al . , machine protection challenges for hl - lhc , proc .
5th international particle accelerator conference , ipac14 , dresden , germany , http://www.jacow.org/. f. zimmermann , m. benedikt , d. schulte and j. wenninger , challenges for highest energy circular colliders , proc .
5th international particle accelerator conference , ipac14 , dresden , germany , http://www.jacow.org/ ; + future circular collider study , https://espace2013.cern.ch/fcc/pages/default.aspx .
f. bordry , status of the large hadron collider ( lhc ) , proc .
2008 european particle accelerator conference , epac08 , genoa , italy , 2008 , http://www.jacow.org/. j. wenninger , status of the lhc , proc .
2009 particle accelerator conference , pac09 , vancouver , canada , 2009 , http://www.jacow.org/. f. bordry et al . , the first long shutdown ( ls1 ) for the lhc , proc .
4th international particle accelerator conference , ipac13 , shanghai , china , 2013 , http://www.jacow.org/. j.ph .
tock et al .
, consolidation of the lhc superconducting circuits : a major step towards 14 tev collisions , proc .
4th international particle accelerator conference , ipac13 , shanghai , china , 2013 , http://www.jacow.org/. r. schmidt et al . , _ new j. phys . _ * 8 * ( 2006 ) 290 .
r. schmidt et al . ,
machine protection and interlock systems for lhc , these proceedings .
appleby et al . , _ phys . rev .
beams _ * 13 * ( 2010 ) 061002 .
v. kain , beam dynamics and beam losses for circular machines , these proceedings .
m. werner et al .
, a fast magnet current change monitor for machine protection in hera , proc .
icaleps 2005 , geneva , switzerland , http://www.jacow.org/. v. kain , beam transfer and machine protection , these proceedings .
b. dehning , lhc beam loss monitor system , these proceedings .
b. auchmann et al . , _ phys . rev .
beams _ * 18 * ( 2015 ) 061002 .
s. redaelli , beam cleaning and collimation systems , these proceedings .
b. salvachua et al . , handling 1 mw losses with the lhc collimation system , proc .
5th international particle accelerator conference , ipac14 , dresden , germany , http://www.jacow.org/. a. bertarelli , beam induced damage mechanisms and their calculation , these proceedings .
v. kain et al .
, material damage test with 450 gev lhc - type beam , lhc - project - report-822 ( 2005 ) .
n. mohkov and f. cerutti , beam material interaction , heating and activation , these proceedings .
gorzawski et al . , the acctesting framework : an extensible framework for accelerator commissioning and systematic testing , proc .
icaleps 2013 , san francisco , ca , usa , http://www.jacow.org/. r. schmidt and j. wenninger , lhc injection scenarios , lhc project note 287 ( 2002 ) .
s. redaelli et al . , operation of the betatron squeeze at the lhc , proc .
4th international particle accelerator conference , ipac13 , shanghai , china , http://www.jacow.org/. t. baer , very fast losses of the circulating lhc beam , their mitigation and machine protection , cern - thesis-2013 - 233 .
t. baer et al . , ufos in the lhc : observations , studies and extrapolations , proc .
3rd international accelerator conference , ipac12 , new orleans , la , usa , http://www.jacow.org/. b. auchmann , blm threshold strategy , lhc performance workshop , chamonix , 2014 , http://indico.cern.ch / event/315665/. j. emery et al . , lhc blm single channel connectivity test using the standard installation , proc .
european workshop on beam diagnostics and instrumentation for particle accelerators dipac09 , basel , switzerland , http://www.jacow.org/. e. ciapala , f. rodriguez mateos , r. schmidt and j. wenninger , the lhc post - mortem system , lhc project note 303 ( 2002 ) .
r. gorbonosov et al .
, plug - in based analysis framework for lhc post - mortem analysis , proc
. international conference on accelerator and large experimental physics control systems icaleps 2013 , san francisco , ca , usa , http://www.jacow.org/. p. charrue et al . , role - based access control for the accelerator control system at cern , proc .
international conference on accelerator and large experimental physics control systems icaleps 2007 , knoxville , tn , usa , http://www.jacow.org/. w. sliwinski et al . , management of critical machine settings for accelerators at cern , proc .
international conference on accelerator and large experimental physics control systems icalepcs 2009 , kobe , japan , http://www.jacow.org/. j. wozniak , v. baggiolini , d. garcia quintas and j. wenninger , software interlock system , proc . international conference on accelerator and large experimental physics control systems icaleps 2007 , knoxville , tn , http://www.jacow.org/. l. ponce , j. wenninger and j. wozniak , operational experience with the lhc software interlock system , proc .
international conference on accelerator and large experimental physics control systems icaleps 2013 , san francisco , ca , usa , http://www.jacow.org/. b. todd , performance and availability of mps 20082012 , proc .
2013 mpp workshop , cern - acc-2014 - 0041 .
r. filippini et al . , reliability analysis of the lhc beam dumping system taking into account the operational experience during lhc run 1 , proc . international conference on accelerator and large experimental physics control systems icaleps 2013 , san francisco , ca , usa , http://www.jacow.org/. j. uythoven , dependability calculations prior to 2008 and operational experience during first lhc run , https://indico.cern.ch/event/277684 .
m. bajko et al . , report of the task force on the incident of 19th september 2008 at the lhc , cern - lhc - project - report-1168 . #
since the lhc powering incident in 2008 was the most severe damage that ever happened to an accelerator , a brief description of it is appended here even though this incident happened _ without beam _
@xcite . in the morning of 19 september 2008 the last commissioning step of the main dipole circuit ( 154 magnets ) of sector 34 ( between ip3 and ip4 in fig .
[ fig : layout ] ) was started , a ramp to 9.3 ka which corresponds to a beam energy of 5.5 tev . during the ramp an electrical fault developed in the powering busbar interconnection at a current of 8.7 ka .
a resistive voltage appeared and increased to 1 v after less than 0.5 s , leading to the power converter trip .
the current started to decrease in the circuit and the energy discharge switch opened , inserting dump resistors in the circuit . in this sequence of events , the quench detection , power converter and energy discharge systems behaved as expected .
no resistive voltage appeared on the dipoles of the circuit , individually equipped with quench detectors and the quench of a magnet has been excluded as initial event . within the first second , a main electrical arc and multiple smaller secondary arcs developed and punctured the helium enclosure , leading to release of helium into the insulation vacuum of the cryostat .
after a few seconds the beam vacuum also degraded . the spring - loaded relief discs on the insulation vacuum enclosure opened when the pressure exceeded atmospheric , thus relieving the helium to the tunnel .
they were however unable to maintain the pressure rise below the nominal 0.15 mpa absolute , thus resulting in large pressure forces acting on the vacuum barriers separating neighbouring subsectors ( a subsector corresponds to two 107 m long cells ) , which damaged them as can be seen in fig .
[ fig : intercon ] .
the forces displaced dipoles in the affected zone from their cold internal supports and knocked the short straight section ( sss ) cryostats housing the quadrupoles and vacuum barriers from their external support jacks at three positions , in some locations breaking the anchors in the concrete floor of the tunnel .
the displacement also damaged the connections to the cryogenic distribution line .
the main damage zone extends over approximately 700 m. about 2 tons of helium were rapidly released to the tunnel , producing a cloud which triggered oxygen deficiency hazard detectors and tripped an emergency stop , thus switching off all electrical power from sector 34 . before restoration of electrical power enabled the actuation of cryogenic valves , another 4 tons of helium were lost at lower flow rates .
the total loss of inventory thus amounts to about 6 tons out of 15 tons initially in the sector .
a post - mortem analysis of cryogenic temperature data revealed a significant temperature anomaly in sector 34 during a powering step to 7 ka performed a few days before the incident . a steady temperature increase of up to 40 mk occurred in the cryogenic cell of the incident .
the excess power in the incident cell corresponds to an unaccounted resistance of around @xmath28 .
given the location of the primary electrical arc , the most likely hypothesis for the cause of the incident is a problem of the busbar joint .
the structure of such a joint is shown in fig .
[ fig : joint ] .
the joints are brazed but not clamped and the nominal joint resistance is @xmath29 .
the incident could be reproduced in simulation assuming a bad electrical and thermal contact of the copper stabilizer at the joint due to lack of solder or poor - quality brazing @xcite .
a total of 53 magnets , 39 dipoles and 14 quadrupole ssss had to be removed from the tunnel and brought to the surface for cleaning and repair .
most of them were replaced with spare magnets .
all magnets have been thoroughly re - tested before re - installation in the tunnel .
both arc beam vacuum chambers were contaminated by soot from electrical arcs and chips of multilayer insulation over roughly 80% of their length as can be seen in fig . [
fig : vac - sooth ] .
contamination by chips of multilayer insulation has been found over long distances away from the position of the original incident .
these chips are deposited mostly on the beam screen surface , from where they are removed by in situ cleaning . following the incident and a review of the magnet and busbar protection system , a quench protection system ( qps )
upgrade was launched to protect all busbar joints of the arc main dipole and main quadrupole circuits .
the required voltage tabs were available , but a large volume of electronics had to be developed .
the busbar protection was finally operational in the spring of 2010 when the lhc was commissioned at 3.5 tev .
the evolution of the lhc beam energy over time is shown in fig .
[ fig : energy - time ] , in 2015 the energy will be pushed to 6.5 tev after the consolidation campaign of the 2 year long shutdown . | since 2010 the large hadron collider ( lhc ) is the accelerator with the highest stored energy per beam , with a record of 140 mj at a beam energy of 4 tev , almost a factor of 50 higher than other accelerators . with such a high stored energy , machine protection aspects set the boundary conditions for operation during all phases of the machine cycle .
only the low - intensity commissioning beams can be considered as relatively safe .
this document discusses the interplay of machine operation and machine protection at the lhc , from commissioning to regular operation .
+ + * keywords * + operation ; machine protection ; beam loss ; lhc . |
the adiabatic - connection fluctuation - dissipation theorem ( acfdt ) provides an exact representation of the electronic correlation energy in term of the interacting density response function , within density functional theory ( dft).@xcite a major advantage of this method , is that it naturally accounts for dispersive interactions through the non - locality of the response function .
furthermore , in contrast to semi - local approximations , the acfdt correlation energy is naturally combined with the exact exchange energy and does not rely on error cancellation between the exchange and correlation contributions to the total energy .
the accuracy of correlation energies within the acfdt , then depends on the quality of the interacting response function which needs to be approximated .
the most famous approximation for the response function is the random phase approximation ( rpa ) , which is obtained when a non - interacting approximation is used for the irreducible polarizability . for metallic systems ,
the rpa provides a qualitative account of screening in and cures the pathological divergence of second order perturbation theory for the homogeneous electron gas . in 2001
furche@xcite applied rpa and acfdt to obtain the dissociation energies of small molecules and found that the results were slightly worse than those obtained with a generalized gradient approximation@xcite ( gga ) with a systematic tendency to underbind .
it was also shown that rpa can account for strong static correlation and correctly reproduces the dissociation limit of the n@xmath0 molecule .
following this , rpa has been applied to calculate cohesive energies of solids@xcite and again , rpa performs significantly worse than gga with a systematic tendency to underbind .
in contrast , rpa produces excellent results for van der waals bonded systems like graphite@xcite , which is very poorly described by semi - local approximations .
in addition , for graphene adsorbed on metal surfaces , where both covalent and dispersive interactions are equally important , the rpa seems to be the only non - fitted scheme capable of describing the potential energy curves correctly.@xcite by now , it is well established that rpa provides a reliable account of van der waals bonded systems but systematically underestimates the strength of covalent and ionic bonds.@xcite furthermore , the absolute correlation energies obtained with rpa are severely underestimated and dissociation energies benefit from huge error cancellations . in particular , for one - electron systems rpa gives rise to a substantial negative correlation energy .
this large self - correlation error can be remedied by subtracting the local rpa error obtained from the homogeneous electron gas,@xcite but unfortunately the procedure does not improve upon dissociation energies of molecules and solids.@xcite a more sophisticated approach is to add a second order screened exchange ( sosex ) contribution to the correlation energy , which exactly cancels the self - correlation energy for one - electron systems .
this approach has been shown to improve dissociation energies of molecules@xcite and cohesive energies of solids@xcite , but is significantly more computationally demanding than rpa .
in addition the sosex term in the correlation energy destroys the good description of static correlation in rpa and produces the wrong dissociation limit of small molecules.@xcite in a different line of development , time - dependent density functional theory@xcite ( tddft ) provides a systematic way to improve the rpa response function . here
the response function can be expressed in terms of a frequency - dependent non - local exchange - correlation kernel and rpa is obtained when the kernel is neglected .
a rather advanced approach in this direction , is the inclusion of the frequency - dependent exact exchange kernel , which has been shown to produce very accurate dissociation energies of small molecules@xcite and conserve the accurate description of static correlation characteristic of rpa.@xcite while this method is considerably more involved than rpa , it provides evidence that accurate correlation energies may be obtained from tddft and acfdt with a good approximation for the exchange kernel . in ref .
[ ] the correlation energy of the homogeneous electron gas was evaluated using different approximations for the exchange - correlation kernel and the results indicated that the frequency dependence of the kernel is of minor importance , while the non - locality of the kernel is crucial .
moreover , it has been shown by furche and van voorhis@xcite that any local approximation for the kernel produces a correlation hole , which diverges at the origin .
the resulting correlation energies then often become worse than those obtained with rpa ( one exception to this is the local energy - optimized kernel of ref .
whereas exchange - correlation kernels have traditionally been derived to produce accurate excited state properties , there is now a considerable interest in obtaining exchange - correlation kernels suited for accurate ground state correlation energies.@xcite in this respect , it is interesting to note that the optical properties of electronic systems are ill described with local approximations for the kernel due to wrong behavior at @xmath1 , while the failure for total energy calculations originate from the bad behavior in the limit @xmath2 ( see discussion below ) . in this paper
we present a parameter - free renormalized adiabatic exchange kernel .
the renormalization introduces non - locality in the kernel and provides an accurate description of the correlation hole at short distances , which gives rise to a better description of short - range correlation compared to rpa .
we note that the philosophy of the renormalization is similar to the smooth cutoff introduced in the energy - optimized kernel of ref .
however , in contrast to that kernel , the present kernel does not contain any fitted parameters .
the kernel has previously been shown to improve upon rpa for absolute correlation energies and dissociation energies of small molecules@xcite , while the computational load is comparable to rpa . here
we describe the theory and implementation in detail and assess the performance for cohesive energies of solids , static correlation and van der waals interactions .
using the adiabatic connection and fluctuation - dissipation theorem ( acfdt ) , the exchange - correlation energy can be written as : @xmath3\big\},\end{aligned}\ ] ] where @xmath4 and @xmath5 is the coulomb interaction . here
@xmath6 is the density , which by definition is constant along the adiabatic connection and @xmath7 is the interacting response function of a system with @xmath8 evaluated at imaginary frequencies .
it is standard practice to divide @xmath9 into an exchange part @xmath10 obtained by setting @xmath11 in the integrand and a correlation part @xmath12 , which is the remainder .
one then obtains @xmath13\big\},\label{e_x}\\ e_{c}&=-\int_0 ^ 1d\lambda\int_0^\infty\frac{d\omega}{2\pi}\text{tr}\big\{v[\chi^\lambda(i\omega)-\chi^{ks}(i\omega)]\big\},\label{e_c}\end{aligned}\ ] ] where @xmath14 is the response function of the non - interacting kohn - sham system .
a major advantage of this separation is that the exchange energy only depends on the occupied bands and can be converged separately with respect to @xmath15-points plane wave cutoff , etc .
one can then focus on the correlation energy @xmath12 , which can be calculated once an approximation for @xmath16 has been given . to obtain @xmath16
we turn to time - dependent density functional theory , where it is easy to show that the interacting response function satisfies the dyson equation @xmath17\chi^\lambda(\omega)\bigg\}.\end{aligned}\ ] ] here , the exchange - correlation kernel @xmath18 is the temporal fourier transform of the functional derivative of the time - dependent exchange - correlation potential at coupling strength @xmath19 .
all the complicated correlation effects contained in @xmath20 has been transferred into @xmath18 , which needs to be approximated . however , even if @xmath18 is neglected all together , one still obtains a non - trivial approximation for @xmath20 due to the coulomb interaction term .
this is the random phase approximation ( rpa ) . to obtain correlation energies beyond the random phase approximation
, we will include an approximation for @xmath18 in eq . .
in general , one can obtain the exchange - correlation kernel along the adiabatic connection from the scaling properties @xcite @xmath21(\mathbf{r},\mathbf{r'},\omega)&=\lambda^2f_{xc}[n'](\lambda\mathbf{r},\lambda\mathbf{r'},\omega/\lambda^2),\\ n'&=\lambda^{-3}n(\mathbf{r}/\lambda),\end{aligned}\ ] ] and it will thus be sufficient to consider the case of @xmath22 . due to the first order nature of exchange ,
any properly derived pure exchange kernel should have the property that @xmath23(\mathbf{r},\mathbf{r'},\omega)=\lambda f_{x}[n](\mathbf{r},\mathbf{r'},\omega)$ ] .
for a pure exchange kernel the coupling constant integration can thus be carried out resulting in @xmath24\notag\\ & + v\chi^{ks}(i\omega)]\big\},\label{e_c_int}\end{aligned}\ ] ] where @xmath25 . however , the inversion of @xmath26 turns out to cause numerical problems and for _ ab initio _ applications in this manuscript we will perform the coupling constant integration numerically .
a simple and natural choice is the adiabatic local density approximation ( alda ) given in the frequency domain by @xmath27(\mathbf{r},\mathbf{r'})&=\delta(\mathbf{r}-\mathbf{r'})f^{\text{alda}}_{xc}[n(\mathbf{r})],\label{alda1}\\ f^{\text{alda}}_{xc}[n(\mathbf{r})]&=\frac{d^2}{dn^2}\big(n\varepsilon_{xc}^{heg}\big)\big|_{n = n(\mathbf{r})},\label{alda2}\end{aligned}\ ] ] where @xmath28 is the exchange - correlation energy per electron in the homogeneous electron gas .
the approximation is in a certain sense similar to lda in static dft , but in contrast to lda , the alda is not exact for the homogeneous electron gas .
in particular , the alda kernel becomes a ( density dependent ) constant in the homogeneous electron gas , while the true kernel should depend on both frequencies and position differences .
this means that alda can violate a number of exact conditions .
for example , it is well known that the kernel in fourier space should behave as @xmath29 for @xmath30 and @xmath29 for @xmath2 , @xcite which are obviously violated by any local kernel .
whereas , the first of these conditions is very important for the description of optical excitations within tddft , it is the second condition , which makes any local approximation @xmath31 useless for total energy calculations in the acfdt framework .
the reason is that the trace in eq . becomes an integral over all @xmath32 and an inaccurate description of the response function at large @xmath33 ,
can deteriorate the correlation energy completely .
equivalently , the constant behavior of @xmath31 at large @xmath33 renders the pair correlation function in real - space divergent at the origin .
this divergence is integrable , but gives rise to severe convergence problems for _ ab initio _
applications.@xcite to show this in more detail we have plotted the correlation hole of the homogeneous electron gas in the top row of fig .
[ fig : g_q ] using rpa and the exchange part of the alda ( alda@xmath34 ) approximation for @xmath31 and compare with an analytic representation.@xcite the analytic representation has been shown to agree with monte carlo simulations and here we will regard this as the exact correlation hole .
the correlation energy per electron is directly related to the integral of the coupling constant averaged correlation hole @xmath35 : @xmath36 it is most instructive to look at the correlation hole in @xmath33-space .
it is seen that the rpa hole decays to slowly and since the correlation energy is proportional to the integral of the correlation hole , is is clear that rpa will overestimate the magnitude of the correlation energy .
in fact , the correlation energy per electron becomes @xmath37 ev too negative for a wide range of densities , whereas the relative error increases from @xmath38 at @xmath39 to @xmath40 at @xmath41.@xcite .
alda@xmath34 , on the other hand , seems to reproduce the @xmath33-space representation of the correlation hole much better .
however , at large @xmath33 , the correlation hole acquires a slowly decaying positive tail and as a consequence , the correlation energy becomes too large by @xmath42 ev per electron .
furthermore , the slowly decaying tail introduces a divergence at the origin in the real space correlation hole .
we note that the results change very little if we include the full alda ( exchange + correlation ) kernel .
if we are mostly concerned with total correlation energies , we may observe that the exact correlation hole picks up most of its weight between zero and @xmath43 where @xmath44 has a zero - point .
the correlation energy is then approximated by @xmath45 and we have previously shown that this procedure yields correlation energies within 30 mev of the exact value for a wide range of densities @xcite . we can now proceed to define a new real space correlation hole obtained by fourier transforming @xmath44 truncated at @xmath43 and we will refer to this as ralda@xmath34 . from fig . [
fig : g_r ] it is evident that this procedure removes the divergence at the origin and gives rise to a correlation hole , which comprises a much better short range approximation than either rpa or alda@xmath34 . at the same time
the ralda@xmath34 correlation hole retains the accurate long - range behavior of the alda@xmath34 correlation hole , which is less accurately described in rpa . in fig .
[ fig : g_r ] we also show the correlation hole weighted by @xmath46 , which is the quantity that should be integrated to obtain the correlation energy .
-space for the homogeneous electrons gas . left : @xmath39 .
right : @xmath47 .
the positive slowly decaying tail of alda is evident at large @xmath33 .
the ralda@xmath34 is obtained by truncating @xmath48 at its zero point at @xmath49,title="fig:",width=4 ] -space for the homogeneous electrons gas .
left : @xmath39 .
right : @xmath47 .
the positive slowly decaying tail of alda is evident at large @xmath33 .
the ralda@xmath34 is obtained by truncating @xmath48 at its zero point at @xmath49,title="fig:",width=4 ] .
right column : @xmath47 .
the bottom row shows the bare correlation hole in real space , where the alda@xmath34 approximation is seen to diverge at the origin .
the ralda@xmath34 is obtained by truncating @xmath50 at @xmath43 ( the zero point of @xmath50 ) and is seen to produce a much better approximation than both rpa and alda@xmath34 .
the top row , shows the correlation hole weighted by @xmath46 , which is the quantity one would integrate to get the correlation energy.,title="fig:",width=4 ] .
right column : @xmath47 .
the bottom row shows the bare correlation hole in real space , where the alda@xmath34 approximation is seen to diverge at the origin .
the ralda@xmath34 is obtained by truncating @xmath50 at @xmath43 ( the zero point of @xmath50 ) and is seen to produce a much better approximation than both rpa and alda@xmath34 .
the top row , shows the correlation hole weighted by @xmath46 , which is the quantity one would integrate to get the correlation energy.,title="fig:",width=4 ] + .
right column : @xmath47 .
the bottom row shows the bare correlation hole in real space , where the alda@xmath34 approximation is seen to diverge at the origin .
the ralda@xmath34 is obtained by truncating @xmath50 at @xmath43 ( the zero point of @xmath50 ) and is seen to produce a much better approximation than both rpa and alda@xmath34 .
the top row , shows the correlation hole weighted by @xmath46 , which is the quantity one would integrate to get the correlation energy.,title="fig:",width=4 ] .
right column : @xmath47 .
the bottom row shows the bare correlation hole in real space , where the alda@xmath34 approximation is seen to diverge at the origin .
the ralda@xmath34 is obtained by truncating @xmath50 at @xmath43 ( the zero point of @xmath50 ) and is seen to produce a much better approximation than both rpa and alda@xmath34 .
the top row , shows the correlation hole weighted by @xmath46 , which is the quantity one would integrate to get the correlation energy.,title="fig:",width=4 ] for the homogeneous electron gas , the cutoff at @xmath43 in @xmath50 can be imposed by using the hartree - exchange kernel @xmath51(q)=\theta\big(2k_f - q\big)f_{hx}^{alda_x}[n].\end{aligned}\ ] ] fourier transforming this expression yields @xmath52(r)&=\widetilde{f}_{x}^{ralda}[n](r)+v^r[n](r),\\
\widetilde{f}_{x}^{ralda}[n](r)&=\frac{f^{alda}_{x}[n]}{2\pi^2r^3}\big[\sin(2k_fr)-2k_fr\cos(2k_fr)\big],\notag\\ v^r[n](r)&=\frac{1}{r}\frac{2}{\pi}\int_0^{2k_fr}\frac{\sin x}{x}dx,\notag\end{aligned}\ ] ] where @xmath53 and we have suppressed the alda exchange ( @xmath54 ) index
. we will refer to this kernel as the renormalized adiabatic local density approximation ( ralda ) .
the cutoff in @xmath33-space is translated into a density dependent width of the delta function in eq . , which gives rise to a non - local exchange - correlation kernel .
similarly , the renormalized hartree kernel acquires a density dependence through the cutoff , but approaches the bare hartree kernel in the limit of @xmath55 .
the kernels are shown in fig .
[ fig : ralda ] .
the kernel has a width which is determined by the density . as @xmath56 ( @xmath57 ) the width decreases and the kernel approaches the bare exchange kernel .
right : the renormalized hartree kernel @xmath58(r)$ ] .
the kernel is finite at the origin and approaches the bare coulomb kernel for large @xmath46.,title="fig:",width=4 ] ( r)$ ] .
the kernel has a width which is determined by the density . as @xmath56 ( @xmath57 ) the width decreases and the kernel approaches the bare exchange kernel .
right : the renormalized hartree kernel @xmath58(r)$ ] .
the kernel is finite at the origin and approaches the bare coulomb kernel for large @xmath46.,title="fig:",width=4 ] interestingly , both the renormalized hartree kernel and exchange kernel become finite at the origin giving @xmath59(r\rightarrow0)&=\frac{4k_f}{\pi}-\frac{8k_f^3r^2}{9\pi},\label{kernel_origin}\\ \widetilde{f}_{x}^{ralda}[n](r\rightarrow0)&=\big[\frac{4k_f^3}{3\pi^2}-\frac{32k_f^5r^2}{15\pi^2}\big]f^{alda}_{x}[n].\end{aligned}\ ] ] this property becomes extremely convenient when the kernel is evaluated in real space .
it is customary to include an exchange - correlation kernel on top of the exact hartree kernel and we thus define the renormalized adiabatic local density approximation by the exchange kernel @xmath60(r)&=\widetilde{f}_{x}^{ralda}[n](r)+v^r[n](r)-v(r).\end{aligned}\ ] ] this representation is also more useful for _ ab initio _ applications to solid state systems since it is difficult to converge the long range tail of @xmath61(r)$ ] .
in contrast , for @xmath55 , [ @xmath61(r)-v(r)]\rightarrow\sin(2k_fr)/r$ ] , which rapidly averages to zero . since @xmath62(r)$ ] decays as @xmath63 for @xmath55 , it is much easier to converge the numerical fourier transform of eq . with respect to sampled unit cells than eq . .
a major advantage of the rpa for _ ab initio _ calculations of total correlation energies , is the fact the spin - polarized systems can be treated by simply making the substitution @xmath64 in eq . .
this is easily shown by using the fact that @xmath65 is independent of spin in rpa , but it no longer holds when a spin - dependent exchange - correlation kernel is used . for exact exchange , one has @xmath66=\frac{e_x[2n_\uparrow]+e_x[2n_\downarrow]}{2},\end{aligned}\ ] ] which translates into @xmath67=2f_x[2n_\sigma]\delta_{\sigma\sigma'}.\end{aligned}\ ] ] here functionals of two arguments are the spinpolarized versions of the spinpaired functionals with one argument .
it is straightforward to impose this spin - scaling on @xmath68 as well , however , we find that this renders the correlation energy very difficult to converge , since the spin - polarized version of the kernel will inherit part of the convergence problems from alda .
this is due to the fact that the spin - diagonal components of the hartree - exchange kernel becomes @xmath69 + 2v^r[2n_\sigma]-v$ ] and the additional bare coulomb interaction @xmath5 destroys the local cancellation of the correlation hole resulting from the spin - density @xmath70 . to obtain the a more useful spin - polarized version of the ralda kernel we first consider the dyson equation with explicit spin dependence : @xmath71\chi_{\sigma''\sigma'},\end{aligned}\ ] ] where we used that the non - interacting response function is diagonal in spin . for a spin - paired system
, we get the constraint that @xmath72=f_x[n].\end{aligned}\ ] ] clearly this is satisfied if we impose the exact condition eq . .
due to the convergence problems mentioned above , we relax eq . and instead use @xmath73=2f^{ralda}_x[n]\delta_{\sigma\sigma'}+v^r[n]-v,\end{aligned}\ ] ] with @xmath74 .
this satisfies eq . and ensures that the renormalized bare coulomb interaction is replaced by the renormalized kernel , when we calculate the full hartree - exchange kernel .
the spin structure is then very similar to that of the alda hartree exchange , the only difference being that we have replaced the bare coulomb interaction with the renormalized coulomb interaction and the alda kernel has been replaced by @xmath75 , where both are evaluated on the total density .
we note that this choice is by no means unique and in our previous work@xcite we used the expression @xmath76\delta_{\sigma\sigma'}+v^r[n_\sigma+n_{\sigma'}]-v$ ] .
however we have found that eq .
gives better results for atomization energies and we have used that approximation in the present work . it is straightforward to generalize the ralda kernel eq . to inhomogeneous systems by taking @xmath77 and @xmath78 .
however , we are forced to introduce a two - point density @xmath79 , which is not uniquely given by the present method .
the only requirement is that @xmath80 which translates into @xmath81 . in the following
we will use two different flavors of the two - point density : @xmath82 and @xmath83 the first of these choices is a simple average of the densities . in our view , this is the most natural choice since the two - point density should give the width of the kernel , if we regard this as a pure function of @xmath84 .
thus if @xmath85 and @xmath86 belong to two well separated systems , the coupling originating from the kernel is determined by the width , which should be given by the average value of the densities at @xmath85 and @xmath86 . from the dyson equation
it is clear that the hartree - exchange - correlation kernel @xmath87 provides a coupling between the non - interacting density response functions @xmath88 and @xmath89 and the kernel thus naturally involves the densities in those points . in the following
we will regard this as the physical choice and implicitly refer to this two - point density unless otherwise stated .
in contrast , @xmath90 is the density at the average position , which becomes zero if @xmath85 and @xmath86 belong to two well separated systems .
this version of the two - point density is therefore not expected to describe van der waals interactions correctly , but it has the great advantage that the kernel becomes a function of @xmath91 and @xmath92 , which simplifies calculations in periodic systems .
nevertheless , _ a priori _ , we would not expect @xmath90 to work well in very inhomogeneous systems .
the renormalized alda functional has been implemented in the dft code gpaw,@xcite which uses the projector augmented wave ( paw ) method.@xcite the response function is calculated in a plane wave basis set as described in refs .
[ ] . we were not able to apply the analytic coupling constant integration eq . due to near singular behavior of @xmath93
. instead we solve the dyson equation for 8 gauss - legendre lambda points and perform the coupling constant integration numerically .
the frequency integration is performed using 16 gauss - legendre points with the highest point situated at 800 ev . the kernel eq .
with a general two - point density like eq .
is only invariant under simultaneous lattice translation in @xmath85 and @xmath86 .
its plane wave representation takes the form @xmath94 where @xmath95 and @xmath96 are reciprocal lattice vectors , @xmath97 is the number of sampled unit cells , @xmath32 belongs to the first brillouin zone , and @xmath98 here we have introduced the lattice point difference @xmath99 and used that each of the @xmath97 sampled unit cell integrals in eq . can be transferred into a single unit cell by letting @xmath100 .
we also used that @xmath101 .
the function @xmath102 is thus periodic in both @xmath85 and @xmath86 and @xmath103 should be converged by sampling a sufficient number of lattice points . for isolated atoms and molecules , one should only use a single term in the sum , but make sure that the calculation is converged with respect to unit cell size . for solids ,
a consistent approach is to sample a set of unit cells , which matches the sampled @xmath15-point grid . since , the sum in eq .
only involves lattice point differences , a lot of terms are equal and we can reduce the double sum over lattice points to a sum over lattice point differences , where each term is weighted by the number of times that particular lattice point difference appears . despite this reduction , the evaluation of the ralda kernel for periodic systems still represents a major computational load , since we have to calculate the full two - point function @xmath104 for each lattice point difference . one way to circumvent the lattice point sampling for periodic systems ,
is to replace the general two - point density eq . by the density at the average position eq . .
the ralda kernel then becomes periodic in the average position and can be written @xmath105(\mathbf{r_-})$ ] , where @xmath106 and @xmath92 . in the limit of @xmath107 the fourier transform in @xmath97 unit cells then becomes , which we replace by @xmath108 for @xmath109 . ]
@xmath110 \theta\big(2k_f[n(\mathbf{r})]-|\mathbf{q}+\mathbf{g_-}|\big)\notag,\end{aligned}\ ] ] where we defined @xmath111 and @xmath112 . in the last line we used the fourier transform of the step function from eq . .
the expression is thus very similar to the fourier transform of the alda kernel except that it involves a density dependent step function .
the most striking improvement of the ralda functional compared to rpa , is the accurate description of absolute correlation energies .
the main motivation for the method is the accurate representation of the correlation energy of the heg and in ref .
[ ] , it was shown that this is also true for simple inhomogeneous systems .
for example , the rpa correlation energy of a hydrogen atom is -0.57 ev , whereas ralda gives -0.02 ev . note that this value differs from our previous work,@xcite due to a different treatment of spin in the present work .
similarly for the h@xmath0 molecule , rpa and ralda yield absolute correlation energies of -2.2 ev and -1.2 ev respectively , which should be compared to the exact value of -1.1 ev .
we have previously demonstrated that the ralda method also significantly improves the accuracy of molecular atomization energies compared to rpa.@xcite in our previous work , the ralda kernel was evaluated using the pseudo - density . in this work
, the ralda kernel is based on the all - electron density and is thus exactly represented within the paw formalism . in this section
we will begin by stating the computational details .
we then present results for atomization energies of molecules and show that the ralda kernel accurately describes strong static correlation in the atomic limit of h@xmath0 dissociation .
cohesive energies of solids are then discussed and the two methods presented in eqs . and are compared .
we then present c@xmath113 coefficients of eight atoms evaluated with lda , rpa and ralda .
finally , we demonstrate that the ralda method becomes very similar to rpa in the description of long range correlation , which we exemplify by the dissociation of a graphene bilayer and the binding energy of four molecular dimers . the calculation of rpa and ralda correlation energies are performed in three steps .
first a standard lda calculation is carried out in a plane wave basis . the full plane wave kohn - sham hamiltonian is diagonalized to obtain all unoccupied electronic states and eigenvalues .
finally , we choose a cutoff energy and calculate the kohn - sham response where we put the number of unoccupied bands equal to the number of plane waves defined by the cutoff and evaluate the correlation energy according to eqs . and .
the calculated correlation energies are added to non - selfconsistent hartree - fock energies evaluated on the same orbitals as the correlation energy . in general
, we have found very little dependence on the input orbitals for hartree - fock energies .
all lda and pbe calculations were performed with a 600 ev cutoff for the wavefunctions . for rpa and ralda energies we use an additional cutoff , which in general is smaller than the wavefunction cutoff and
defines the number of plane wave used to represent the response function .
the calculations are challenging to converge with respect to this cutoff energy and for most calculations we have used an extrapolation scheme to obtain converged results . for sufficiently high cutoff energies the correlation energy is found to scale as@xcite @xmath114 where @xmath115 is the converged result . for most
considered systems the extrapolation was found to be very accurate when the cutoff is increased beyond 300 ev .
typically , the converged results are obtained by extrapolating correlation energies in the range 250 - 400 ev .
when comparing systems with similar electronic structure in the same unit cell , energy differences converge much faster and in the case of a bilayer graphene we used a fixed cutoff of 200 ev to calculate the potential energy curve .
the ralda kernel is evaluated on a real space grid where the grid spacing is set to @xmath116 .
we have checked that ralda results are converged with respect to this grid spacing which is always on the order @xmath117 when the wavefunction cutoff energy is 600 ev ( see appendix a ) .
all calculations were performed on experimental geometries corrected for zero - point anharmonic effects .
for the small atoms and molecules , the simulation was carried out in a periodic supercell where the shortest distance to an atom in a neighboring cell is 6 . for si ,
ge , mg , li , al , pd , rh , cu , ag we used 8 and for na we used 10 .
all solids were simulated with a 12x12x12 gamma - centered @xmath15-point sampling for the correlation energy calculations and a fermi - dirac smearing of 0.01 ev .
the hartree - fock energies were found to converge much slower with respect to @xmath15-point sampling and we typically used 18 - 22 @xmath15-points in each direction for this . for bilayer graphene we used a @xmath15-point sampling of 16x16 for both hartree - fock and correlation energies .
.atomization energies of diatomic molecules .
the alda values are taken from ref .
[ ] and experimental values ( corrected for zero point vibrational energies ) are taken from ref .
all number are in kcal / mol .
the bottom line shows the mean absolute error for this small test set . [ cols="^,^,^,^,^,^,^,^,^",options="header " , ] in fig .
[ fig : bilayer ] we show the binding energy curve of an a - b stacked bilayer of graphene . as in the case of graphite , the pbe functional predicts a very weak binding energy and an equilibrium distance of 4.44 . the van der waals density functional of ref .
[ ] gives a binding energy of 22 mev and an equilibrium distance of 3.65 .
rpa gives a binding energy of 25 mev and an equilibrium distance of 3.39 , whereas ralda gives a binding energy of 22 mev and an equilibrium distance of 3.45 .
we should note that the binding energy curves were obtained by fitting a rather rough set of interlayer distances and more accurate results would require more accurate sampling of the binding energy curve .
the rpa and ralda curves coincide a large distances and decay slower than the tail of the curve obtained with the van der waals functional . in general rpa
has a tendency to underbind and the reduced binding energy of ralda compared to rpa could imply that ralda actually performs worse than rpa in this case .
however , in the case of graphite , rpa seems to produce the exact binding energy between layers@xcite and it is not clear if rpa is also expected to underbind in this case .
furthermore , rpa and ralda seems to give identical results for the few s22 dimers and while we do not have accurate experimental data for the binding energy and equilibrium distance of bilayer graphene , fig .
[ fig : bilayer ] merely serves to illustrate the applicability and similarity of rpa and ralda for the description of long range dispersive interactions .
we have assessed the performance of a non - local adiabatic kernel ( ralda ) for electronic correlation energies . for atomization energies of small molecules and cohesive energies of insulators and metals , the kernel performs significantly better than rpa .
we have found inclusion of semi - core states is crucial for the description of transition metals at the rpa and ralda level .
for the small molecules , we also obtain better agreement with experiments than the sosex extension to rpa .
the method preserves ( but does not improve ) the good description of static correlation and dispersive interactions , which is the main merit of rpa .
the kernel also improves the description of c@xmath113 coefficients compared to rpa . in our opinion ,
a major advantage of the present approach is the unique choice of input orbitals and eigenvalues , which should match the adiabatic ( renormalized ) kernel . while lda orbitals may not be the most accurate choice for evaluating the kohn - sham response function , it is satisfactory that we do not have to make an arbitrary choice , which can have a severe influence on the results . from the point of view of tddft
, rpa corresponds to the time - dependent hartree approximation and a consistent choice in that case would be to adopt hartree orbitals and eigenvalues
. however , this would be a very bad starting point and rpa calculations are usually performed on top of semilocal dft orbitals or hartree - fock orbitals . in this sense , the non - selfconsistent ralda calculations represent a more complete and consistent scheme than non - selfconsistent rpa .
the computational cost of the method is larger than rpa , but certainly cheaper than time - dependent exact exchange or sosex .
the main differences compared to rpa is the evaluation of the renormalized non - local kernel . for non - periodic systems
this evaluation does not comprise a major extension of the computational time compared to rpa .
for the cohesive energies of solids a large unit cell sampling is required to converge results and the calculations are somewhat more time - consuming than rpa . on the other hand for bilayer graphene , the potential energy surfaces are converged with a unit cell sampling of 4 adjacent cell and the computational cost of ralda is just 20 % extra compared to rpa .
the cohesive energies of solids are important for an assessment of the method , but for future application to larger systems ( for example molecular adsorption or reaction barriers at metal surfaces ) the calculations will be dominated by the evaluation of the repsonse function and the computational cost becomes comparable to rpa . compared to rpa , another complication is the spin - dependence of the kernel , which forces us to solve the full spin - dependent dyson equation instead of working with spin - summed quantities . in our plane wave implementation , this leads to memory problems for spin - polarized systems when considering large unit cells and high cutoff energies . finally , the coupling constant integration is carried out numerically due to near - singular behavior of the kernel . however , this problem is very likely to originate from the plane - wave representation of the kernel and a different implementation could probably solve this problem and apply the analytic result . in the present paper
we have only discussed the renormalized alda kernel .
however , it should be straightforward to generalize this to a renormalized adiabatic pbe kernel or in fact , any semi - local exchange kernel
. we would expect this to improve results further due to more accurate initial orbitals and eigenvalues and because the kernel would then contain gradient corrections , which are likely to improve the description of the interacting response function .
the cutoff scheme thus implies an entire hierarchy of renormalized adiabatic kernels for correlation energy calculations .
one caveat , is the fact that many semi - local approximations to the exchange correlation energy work well due to error cancellation between exchange and correlation and it is possible that one would have to include the correlation part of the kernel as well , in order to obtain highly accurate results .
this would , however , be straightforward and one would simply have to evaluate the full kernel explicitly along the adiabatic connection since the linear scaling of pure exchange is lost .
it would also be very interesting to compare the present results , with those obtained with the fitted non - local kernel of ref .
@xcite , which is defined by a similar cuf off procedure .
we will leave these issues to future work .
the two - point kernel eq .
is evaluated using a density , which is represented on a real space grid .
we have not been able to derive an explicit paw correction for the kernel , and have used the all - electron density for calculations in this work . since the all - electron density varies rapidly in the core region of atoms , it is not obvious that it is possible to converge the kernel with respect to grid spacing .
however , the alda exchange kernel approaches zero for large densities and as it turns out , the rapidly oscillating core region does not contribute much to the ralda kernel .
this is illustarted in fig .
[ fig : n2_grid ] where the correlation energy of an n@xmath0 molecule is plotted for decreasing grid spacing .
the energy difference ( contribution to the atomization energy ) converges rapidly and is accurate to within 10 mev at 0.17 .
this is the value used for all calculations in the present work and is close to the default value of 0.18 in gpaw .
the correlation energy of a single n atom is also close to convergence at this value , whereas a slightly smaller grid spacing is required for n@xmath0 . for the molecular and atomic systems in this work we have obtained the plane wave representation of the ralda kernel by evaluating the full hartree - exchange kernel in real space and performing a numerical fourier transform @xmath118(\mathbf{r},\mathbf{r}')+v^r[n](\mathbf{r},\mathbf{r}')\big].\ ] ] in a single unit cell .
this results in a truncation on the kernel whenever @xmath85 or @xmath86 is outside the unit cell , which means that the system will not interact with its periodic images .
another major advantage of this approach is that the full hartree - exchange kernel is finite for @xmath119 eq . , which is not true for the bare exchange kernel eq . . for solids , however , it is important to take into account the long range nature of the coulomb interaction .
the renormalized coulomb interaction @xmath120 approaches the bare coulomb interaction when @xmath55 and it becomes difficult to converge the fourier transform of this with respect to the number of sampled unit cells . for periodic systems , we therefore only represent the exchange part of the kernel in real space and add the exact fourier transform of the hartree kernel : @xmath121,\end{aligned}\ ] ] where @xmath122 .
this representation is easier to converge since @xmath123r)/r$ ] for @xmath55 .
this is illustrated in fig .
[ fig : pd_kpts ] where we compare the convergence of the two implementations with respect to the number of sampled unit cells .
it is seen that eq .
exhibits convergence behavior very similar to rpa , which means that the convergence is largely governed by brillouin zone sampling of kohn - sham states .
in contrast , the implementation eq .
converges as @xmath124 where @xmath97 is the number of sampled unit cells .
we also show the difference between the two methods , which slowly approaches zero in the limit of @xmath107
. however , the bare exchange kernel diverges for @xmath119 , and this divergence is represented by a spherical average around a single grid point , which make the calculations converge slightly slower with respect to grid spacing .
the authors acknowledge support from the danish council for independent research s sapere audie program , grant no 11 - 1051390 .
the center for nanostructured graphene is sponsored by the danish national research foundation . | we assess the performance of a recently proposed renormalized adiabatic local density approximation ( ralda ) for _ ab initio _ calculations of electronic correlation energies in solids and molecules .
the method is an extension of the random phase approximation ( rpa ) derived from time - dependent density functional theory and the adiabatic connection fluctuation - dissipation theorem and contains no fitted parameters .
the new kernel is shown to preserve the accurate description of dispersive interactions from rpa while significantly improving the description of short range correlation in molecules , insulators , and metals . for molecular atomization energies
the ralda is a factor of 7(4 ) better than rpa(pbe ) when compared to experiments , and a factor of 3(1.5 ) better than rpa(pbe ) for cohesive energies of solids . for transition metals the inclusion of full shell semi - core states
is found to be crucial for both rpa and ralda calculations and can improve the cohesive energies by up to 0.4 ev .
finally we discuss straightforward generalizations of the method , which might improve results even further . |
last decade has seen a major shift in stellar spectroscopy : a slow collection of individual spectra has been accelerated by massive surveys , mostly using fiber - fed spectrographs with hundreds of spectra observed simultaneously .
the past and ongoing efforts include rave @xcite , gaia - eso @xcite , segue @xcite , apogee @xcite , lamost @xcite , galah @xcite , and of course gaia @xcite .
up - to - date overviews of the state and results of these surveys are given elsewhere in this volume .
the main goal of stellar spectroscopic surveys is to study galactic structure and evolution . but the collected spectra allow for a significant auxiliary science .
the three examples discussed below are an illustration of a vast range of posibilities and are by no means exhaustive .
we believe that every observer could add further relevant uses of hundreds of thousands of stellar spectra , which were in most cases selected for observation only following simple positional and magnitude constraints .
the first example illustrates research of the multi - dimensional structure of the interstellar medium .
the next one helps with identifying young stars in the field .
the last one is an example on how even a single spectrum obtained by a stellar survey can improve the solution of an astrometric binary which is being derived by gaia .
in 2020 , the gaia mission ( launched in december 2013 ) is expected to release 6-dimensional ( spatial position + velocity ) vectors for a significant fraction of stars on our side of the galactic centre , thus allowing a computation of stellar orbits and of evolution of the galaxy as a whole .
traditional studies of the galactic interstellar medium ( ism ) can not yield information equivalent to stars , as absorption studies get only a 2-dimensional ( column density ) information by observing one hot star at a time .
but ism allows to open up its 3-rd and 4-th dimension by studying diffuse interstellar bands ( dibs ) , weak but numerous absorption lines seen in spectra of background stars which are likely caused by distinct macromolecular carriers .
high dimensionality requires measurement of the strength of these weak interstellar lines also for cool stars which by far outnumber hot stars in the galaxy .
recent new approaches divide out the cool star spectrum by use of synthetic models of stellar atmospheres @xcite or in a self - calibrated way by using spectra of similar stars with negligible ism absorption observed at high galactic latitudes by the same survey @xcite . by observing a given dib toward
many stars which are nearly in the same direction but at different and known distances one can reconstruct absorption sites along the line of sight . joining observations in many directions on the sky
then gives their spatial distribution .
finally , measurement of radial velocity shift yields a 4-dimensional picture of the ism for each dib , and can even constrain placement of multiple clouds along each line of sight .
interstellar absorption lines of sodium and potassium atoms yield information equivalent to dibs , but emission lines or dust absorptions are limited to up to 3 dimensions .
ism is the place of violent collisions of supernova shells , plus winds from asymptotic giant branch stars and hot - star associations .
head - on collisions in the galactic plane are difficult to interpret , though an expected galactic rotation pattern has been nicely identified @xcite . but
observations of the on - going galah and partly gaia - eso surveys are away from the plane where interactions generally result in a net motion perpendicular to the plane .
if any shells of absorbing material are identified we can assume that their motion is perpendicular to shell surfaces and reconstruct a complete velocity vector from its radial velocity component .
such information for ism is then equivalent to the one collected for stars by gaia .
this information can be used to study past events in the interstellar medium . @xcite
published a quasi 3-dimensional map of intensity of diffuse interstellar band at 8620 which shows that distribution of dib extinction is thicker than the one of dust and that it is different on either side of the galactic plane , a witness to asymmetries in placement of recent explosions of supernovae and to incomplete vertical mixing .
observations with the gaia - eso and galah surveys could be used to increase the dimensionality of ism studies to 4 dimensions ( for an example of radial velocity measurements see * ? ? ?
they could also identify and characterize galactic fountains blown away by supernovae in the last million years .
such flows are thought to sustain star formation in the disk by entraining fresh gas from the halo , so they provide a mechanism which explains why star formation in our and other similar galaxies did not stop when gas present in the disk has been used up @xcite .
figure [ figdibsgalah ] plots a dozen dibs and the k i interstellar atomic line at 7699 in a stellar spectrum observed by galah .
spectrum of tyc 4011 - 102 - 1 , a hot star with strong interstellar absorptions close to the galactic plane , is shown .
each 20 wide panel is centred on the dib wavelength as listed in @xcite .
plotted wavelengths are heliocentric .
right - most panel identifies two interstellar clouds for k i at different velocities . for a majority of galah objects , which lie away from the galactic plane , such complications are rare ( but
can be detected ) .
properties of a star are entirely determined by its initial composition , mass and current age if one neglects rotation , magnetism or multiplicity . as noted by david @xcite ``
age is not a direct agent of change and can not be measured like mass or composition .
also , age affects the core of the star , but we observe the surface which is complex . ''
large spectroscopic surveys have the possibility to measure some empirical age indicators , i.e. rotation , activity , and lithium depletion boundary .
the galah survey will bring studies of these age indicators to industrial scale with its hundreds of thousands of observed spectra .
lithium depletion studies have been motivated by lithium observations of main - sequence stars in young clusters and the halo ( e.g. * ? ? ?
the galah survey includes the li i 6708 line in its red channel .
its resolving power of @xmath2 and a typical s@xmath3n ratio of 100 per resolution element allow for efficient measurement of stellar rotation and for studies of profiles of h@xmath4 and h@xmath5 lines which are sensitive to chromospheric activity .
measurement of these youth indicators for field stars is important , as it may point to stars recently ejected from young stellar environments .
parallaxes and proper motions measured by gaia , together with spectroscopically derived radial velocities permit to reconstruct their galactic orbits and so to identify recently dispersed stellar clusters .
multidimensional chemistry studies , which are within the scope of galah , are then the final check of the emerging picture based on chemical tagging @xcite .
stellar activity identification is now entering the era of massive studies .
figure [ figactivityrave ] summarizes active star candidates found in rave data using equivalent width of the emission components of the ca ii infrared triplet lines ( ew@xmath6 , * ? ? ?
grey histogram is a distribution of ew@xmath6 for stars with active morphology in rave , as identified by a locally linear embedding technique @xcite .
solid line marks normal stars which are assumed to be inactive , while dashed line marks rave stars classified by simbad to be pre - main sequence stars .
@xmath7 is a logarithmic measure for the probability that a star with a given ew@xmath6 differs from an inactive spectrum .
its values from left to right correspond to the probabilities of 5 and 2 @xmath8 below zero , zero , and 2 , 5 and 10 @xmath8 above zero .
altogether the work identifies @xmath9 stars with chromospheric flux in ca ii lines detectable at least at a 2 @xmath8 confidence level . [ tablefractions ] crrrrrrr prevalent & + category & @xmath10&@xmath11&@xmath12&@xmath13&@xmath14&@xmath15&@xmath16 + @xmath10 & 80.2 & 0.8 & 3.0 & 0.4 & 0.5&13.5 & 1.5 + @xmath11 & 1.4&80.0 & 0.1 & 0.5 & 4.0 & 3.6&10.4 + @xmath12 & 4.0 & 0.1&73.6 & 1.7 & 1.3&18.6 & 0.6 + @xmath13 & 1.9 & 9.7 & 1.0&64.0 & 3.5 & 3.3&16.7 + @xmath14 & 1.4 & 5.2 & 0.1 & 0.2&77.2 & 5.3&10.5
+ @xmath15 & 12.1 & 2.5&13.2 & 1.5 & 1.8&67.1 & 1.9 + @xmath16 & 0.3 & 4.3 & 0.1 & 1.7 & 5.3 & 1.6&86.7 + presence of emission components in the ca ii infrared triplet ( rave , gaia , and gaia - eso ) or in balmer lines ( galah and gaia - eso ) do not prove that the object is young : interacting binaries are an obvious example of old objects with emission type spectra .
but such objects are not very common in the field .
rave ( fig .
[ figactivityrave ] ) found that strong emissions suggest a pre main - sequence evolutionary phase .
this is consistent with results of the gaia - eso survey , where @xcite studied 22,035 spectra of stars in young open cluster fields and found that 7698 spectra ( 35% ) belonging to 3765 stars have intrinsic emission in h@xmath4 .
again , such a large fraction of emission type spectra in a young stellar environment suggests that emission is related to youth .
but emission is a transient property and morphological classification of emission may be changing with time .
@xcite shows that most profiles are composed and classifies such profiles by properties of fits using two gaussians : @xmath10 stands for blended emission components , @xmath11 have double sharp peaks , @xmath12 are double emission , @xmath13 are p - cygni profiles , @xmath14 are inverted p - cygni , @xmath15 is self - absorption and @xmath16 is emission within absorption .
off - diagonal elements in table [ tablefractions ] report correlations betweeen individual composed profile types .
when emission blend is the prevalent category for an object , it is most often in combination with self - absorption , which is best explained by one of the two components being in transition between absorption and emission .
similarly , the double sharp peaks can change to emission with absorption if a relatively weak absorption is constantly present and one of the emission peaks diminishes .
the largest off - diagonal elements connect double peaks and self - absorption .
the distinction between the two categories is largely influenced by the inclination of the slopes in the profile that are liable to change in the presence of additional weaker components or they are harder to retrieve in the case of more noisy spectra .
the most frequently identified morphological categories from @xcite are emission blend ( 1729 spectra ) , emission in absorption ( 1652 spectra ) , and self absorption ( 1253 spectra ) .
we conclude that many stars have their emission transient in time or in morphological type , so that activity detected through emission is an indication of youth which is not always present and should be used in connection with the absolute position of the star on the h - r diagram , a frequently known property in the gaia era .
gaia will observe huge numbers of different types of binaries @xcite and study them with a wide range of techniques ( fig .
[ figbinaries ] ) .
one of its core strengths will be a derivation of accurate astrometric solutions even for binaries with extreme mass ratios @xcite .
on the other hand spectroscopy from ground based surveys will be the source of detailed chemistry for any type of binary or multiple system .
astrometry has been frequently used to supplement spectroscopic observations in the past ( e.g. * ? ? ?
* ) , but in gaia the opposite will be a common case ( e.g. * ? ? ? * ) .
many astrometric binaries will have components of similar mass and luminosity .
the reach of astrometry is limited in this case : the two stellar images are usually not spatially resolved , so that gaia will be able to trace only the astrometric motion of the photocenter of the two components .
such studies yield accurate orbital period , but since the photocenter is located somewhere between the two stars individual masses can not be derived from astrometry alone . here
even a single spectrum obtained during a spectroscopic survey can be extremely valuable .
radial velocities of individual components in an sb2 at an orbital phase known from astrometry allow to derive the true sizes of both orbits , and so the complete solution of the system .
a proper bayesian analysis of simultaneous astrometric and spectroscopic information will be needed for this task @xcite .
bland - hawthorn , j. 2009 , in the galaxy disk in cosmological context , iau symp .
254 , 241 de silva , g. , et al .
2015 , , 449 , 2604 eyer , f. , et al .
2015 , in living together : planets , host stars and binaries , asp conf .
, arxiv:1502.03829 fraternali , f. 2014 . in setting the scene for gaia and lamost , iau symp
298 , 228 freeman , k. , bland - hawthorn , j. 2002 , , 40 , 487 gilmore , g. , et al .
2012 , the messenger , 147 , 25 jenniskens , p. , dsert , f .- x .
1994 , a&as , 106 , 39 kordopatis , g. , et al .
2013 , , 146 , 134 kos , j. , et al .
2013 , , 778 , 86 kos , j. , et al .
2014 , science , 345 , 791 kos , j. 2015 , this volume luo , a .-
2015 , arxiv e - prints:1505.01570v1 matijevi , g. , et al .
2012 , , 200 , 14 prusti , t. 2014 , in asteroids , comets , meteors 2014 puspitarini , l. , et al .
2015 , , 573 , 35 sahlmann , j. , fekel , f.c .
2013 , , 556 , a145 steinmetz , m. , et al .
2006 , , 132 , 1645 schulze - hartung , t. , launhardt , r. , henning , t. 2012 , , 545 , a79 siebert , a. , et al .
2011 , , 141 , 187 soderblom , d. 1995 , mem .
italiana , 66 , 347 soderblom , d. 2014 , in astrophysical calibration of gaia and other surveys , ringberg castle , 7 - 11 july 2014 , http://www2.mpia-hd.mpg.de/gaia/gaiacal2014/ sderhjelm , s. 2004 , in spectroscopically and spatially resolving the components of the close binary stars , asp conf .
, 318 , 413 torres , g. 2006 , , 131 , 1022 traven , g. , et al .
2015 , , submitted yanny , b. , et al .
2009 , , 137 , 4377 zasowski , g. , et al .
2013 , , 146 , 81 zasowski , g. , et al .
2015 , apj , 798 , 35 zwitter , t. , munari , u. 2004 , in the environment and evolution of double and multiple stars , iau coll . 191 , rev . mex .
conf . , 21 , 251 zwitter , t. , et al . 2008 , , 136 , 421 erjal , m. , et al . 2013 , , 776 , 127 | current ongoing stellar spectroscopic surveys ( rave , galah , gaia - eso , lamost , apogee , gaia ) are mostly devoted to studying galactic archaeology and structure of the galaxy .
but they allow for important auxiliary science :
( i ) galactic interstellar medium can be studied in four dimensions ( position in space @xmath0 radial velocity ) through weak but numerous diffuse insterstellar bands and atomic absorptions seen in spectra of background stars , ( ii ) emission spectra which are quite frequent even in field stars can serve as a good indicator of their youth , pointing e.g. to stars recently ejected from young stellar environments , ( iii ) astrometric solution of the photocenter of a binary to be obtained by gaia can yield accurate masses when joined by spectroscopic information obtained serendipitously during a survey .
these points are illustrated by first results from the first three surveys mentioned above .
these hint at the near future : spectroscopic studies of the dynamics of the interstellar medium can identify and quantify galactic fountains which may sustain star formation in the disk by entraining fresh gas from the halo ; rave already provided a list of @xmath1 14,000 field stars with chromosperic emission in ca ii lines , to be supplemented by many more observations by gaia in the same band , and by galah and gaia - eso observations of balmer lines ; several millions of astrometric binaries with periods up to a few years which are being observed by gaia can yield accurate masses when supplemented with measurements from only a few high - quality ground based spectra . |
many of the high - energy theories of fundamental physics are formulated in higher - dimensional spacetimes . in particular , the idea of extra dimensions has been extensively used in supergravity and superstring theories .
it is commonly assumed that the extra dimensions are compactified . from the inflationary point of view , universes with compact spatial dimensions , under certain conditions , should be considered a rule rather than an exception @xcite .
models involving a compact universe with non - trivial topology play a very important role by providing proper initial conditions for inflation . and compactification of spatial dimensions leads to a number of interesting quantum field theoretical effects , which include instabilities in interacting field theories , topological mass generation , and symmetry breaking . in the case of non - trivial topology ,
the boundary conditions imposed on fields give rise to a modification of the spectrum for vacuum fluctuations and , as a result , to casimir - type contributions in the vacuum expectation values of physical observables ( for the topological casimir effect and its role in cosmology see @xcite and references therein ) . in models of the kaluza - klein type
, the casimir effect has been used as a stabilization mechanism for moduli fields and as a source for dynamical compactification of the extra dimensions , in particular , for quantum kaluza - klein gravity ( see ref .
@xcite ) . the casimir energy can also serve as a model for dark energy needed for the explanation of the present accelerated expansion of the universe ( see @xcite and references therein ) .
in addition , recent measurements of the casimir forces between macroscopic bodies provide a sensitive test for constraining the parameters of long - range interactions , as predicted by modern unification theories of fundamental interactions @xcite .
the influence of extra compactified dimensions on the casimir effect in the classical configuration of two parallel plates has been recently discussed in @xcite-@xcite , for the case of a massless scalar field with dirichlet boundary conditions , and in @xcite-@xcite , for the electromagnetic field for perfectly conducting boundary conditions .
more recently , interest has concentrated on the topic of the casimir effect in braneworld models with large extra dimensions . in this type of models ( for a review see @xcite ) the concept of brane is used as a submanifold embedded in a higher dimensional spacetime , on which the standard - model particles are confined .
braneworlds naturally appear in the string / m theory context and provide a novel set up for discussing phenomenological and cosmological issues related with extra dimensions . in braneworld models
the investigation of quantum effects is of considerable phenomenological interest , both in particle physics and in cosmology .
the braneworld corresponds to a manifold with boundaries .
all fields which propagate in the bulk will give casimir - type contributions to the vacuum energy and , as a result , to the vacuum forces acting on the branes .
casimir forces provide a natural mechanism for stabilizing the radion field in the randall - sundrum model , as required for a complete solution of the hierarchy problem .
in addition , the casimir energy gives a contribution to both the brane and the bulk cosmological constants .
hence , it has to be taken into account in any self - consistent formulation of the braneworld dynamics . the casimir energy and corresponding casimir forces within the framework of the randall - sundrum braneworld @xcite
have been evaluated in refs .
@xcite-@xcite by using both dimensional and zeta function regularization methods .
local casimir densities were considered in refs .
the casimir effect in higher dimensional generalizations of the randall - sundrum model with compact internal spaces has been investigated in @xcite-@xcite .
the purpose of the present paper is to study the casimir energy and force for a massive scalar field with an arbitrary curvature coupling parameter , obeying robin boundary conditions on two codimension one parallel plates which are embedded in the background spacetime @xmath1 , being @xmath2 an arbitrary compact internal space .
the most general case is considered , where the constants in the boundary conditions are different for the two separate plates .
it will be shown that robin boundary conditions with different coefficients are necessary to obtain repulsive casimir forces .
robin type conditions are an extension of dirichlet and neumann boundary conditions and genuinely appear in a variety of situations , including vacuum effects for a confined charged scalar field in external fields @xcite , spinor and gauge field theories , quantum gravity and supergravity @xcite .
robin conditions can be made conformally invariant , while purely - neumann conditions can not .
therefore , robin type conditions are needed when one deals with conformally invariant theories in the presence of boundaries and wishes to preserve this invariance .
it is interesting to note that a quantum scalar field satisfying robin conditions on the boundary of a cavity violates the bekenstein s entropy - to - energy bound near certain points in the space of the parameter defining the boundary conditions @xcite .
robin boundary conditions are an extension of those imposed on perfectly conducting boundaries and may , in some geometries , be useful for modelling the finite penetration of the field through the boundary , the skin - depth parameter being related to the robin coefficient @xcite . in other words ,
those are the boundary conditions which are more suitable to describe physically realistic situations .
this type of boundary conditions naturally arise for scalar and fermion bulk fields in the randall - sundrum model @xcite and the corresponding robin coefficients are related to the curvature scale and to the boundary mass terms of the field .
robin boundary conditions also appear in the study of casimir forces between the boundary planes of films ( for a recent discussion with references see , for instance , @xcite ) .
the casimir effect in the geometry of two parallel plates with robin boundary condition was investigated in refs .
note moreover that boundary problems with non - local boundary conditions can also be reduced to corresponding ones with robin conditions , with the coefficients depending on the wave vector components along the plates @xcite .
the outline of the paper is as follows . in the next section
we will consider the geometry of the problem and the corresponding eigenfunctions .
the casimir energy for two parallel plates in the general case for the internal subspace is evaluated in sect .
[ sec : casen ] .
the boundary - free and single plate parts will be extracted in a cutoff independent way .
applications to braneworlds are then discussed . in sect .
[ sec : casforce ] we consider the casimir forces and show that depending on the coefficients in the boundary conditions these forces can be either attractive or repulsive .
the asymptotic behavior of the forces , for small and large interplate distances , is given .
as an application of the general results , in sect .
[ sec : example ] a simple example with an internal space @xmath3 is discussed , for general periodicity condition along the compactified dimension . for this special example , we also present the boundary - free part and extract from the single plate parts the topological contributions .
the corresponding generalizations for the internal spaces @xmath4 and @xmath5 are also given in detail . in sect .
[ sec : piston ] we extend the results for the casimir energy and force to the case of the geometry of a piston with arbitrary cross - section .
section [ sec : conc ] contains a summary of the work .
we consider a scalar field @xmath6 , with arbitrary curvature coupling parameter @xmath7 , satisfying the equation of motion @xmath8@xmath9 , with @xmath10 being the scalar curvature for a @xmath0-dimensional background spacetime ( for the metric signature and the curvature tensor we adopt the conventions of ref .
@xcite ) . for the special cases of minimally and of conformally coupled scalars
one has , respectively , @xmath11 and @xmath12 .
we will assume that the background spacetime has a topology @xmath13 , where @xmath14 is @xmath15-dimensional minkowski spacetime and @xmath2 a @xmath16-dimensional internal manifold , @xmath17 .
the corresponding line element has the form @xmath18with @xmath19 being the metric for the @xmath15-dimensional minkowski spacetime and the coordinates @xmath20 cover the manifold @xmath2 . here and below @xmath21 and @xmath22 . for the scalar curvature of the metric tensor , from ( [ metric ] ) one has @xmath23 , where @xmath24 is the scalar curvature for the metric tensor @xmath25 .
our main interests in this paper will be to study the casimir energy density and the mutual forces occurring for the geometry of two parallel infinite plates of codimension one , located at @xmath26 and @xmath27 , @xmath28 .
as most general set up , we assume that on these boundaries the scalar field obeys robin boundary conditions @xmath29\varphi ( x)=0,\quad x^{d_{1}}=a_{j},\;j=1,2 , \label{boundcond}\]]with constant coefficients @xmath30 . for @xmath31
these boundary conditions are reduced to dirichlet one and for @xmath32 to neumann boundary conditions .
the choice of different boundary conditions on the plates may correspond physically to use of different materials for plates .
the imposition of boundary conditions on the quantum field changes the spectrum for the zero point fluctuations and leads to the modification of the vacuum expectation values for physical quantities , as compared with the same situation without boundaries . in the region between the plates , @xmath33 , the corresponding eigenfunctions , satisfying the boundary condition on the plate at @xmath34 ,
can be expressed in the decomposed form : @xmath35\ , \psi _
{ \beta } ( x ) , \label{eigfunc1}\]]where @xmath36 denotes a set of quantum numbers specifying the solution and @xmath37 in eq .
( [ eigfunc1 ] ) , the @xmath38 , @xmath39 , are defined by the relations @xmath40 the modes @xmath41 are the eigenfunctions of the operator @xmath42 : @xmath43 \psi _ { \beta } ( x)=-\lambda _ { \beta } ^{2}\psi _ { \beta } ( x ) , \label{eqint1}\]]with eigenvalues @xmath44 , and fulfill the normalization condition @xmath45 in eq .
( [ eqint1 ] ) , @xmath46 is the laplace - beltrami operator for the metric @xmath25 . in the consideration below
we will assume that @xmath47 .
from the boundary condition on the second plate one obtains that the eigenvalues for @xmath48 are solutions of the equation@xmath49we denote by @xmath50 , @xmath51 , the zeros of the function @xmath52 in the right half - plane of the complex variable @xmath53 , arranged in ascending order , @xmath54 . in the discussion below
we will assume that all these zeros are real .
this is the case for the conditions ( see rome02 ) @xmath55 .
the coefficient @xmath56 in ( [ eigfunc1 ] ) is determined from the orthonormality condition for the eigenfunctions , and is equal to@xmath57 ^{-1 } , \label{c2}\]]being @xmath58 the eigenfrequencies .
the vacuum energy in the region between the plates ( per unit volume along the directions @xmath59 ) is given by the formal expression@xmath60}=\frac{1}{2}\int \frac{d\mathbf{k}}{(2\pi ) ^{d_{1}-1}}% \sum_{\beta } \sum_{n=1}^{\infty } \sqrt{k^{2}+z_{n}^{2}/a^{2}+m_{\beta } ^{2}}. \label{elr}\]]in the discussion below we will assume that some cutoff function is present , without writing it explicitly .
alternatively , one can use zeta function regularization , that yields the same result . for the sum over @xmath61 we use the summation formula
@xcite@xmath62by taking into account the relation@xmath63we see that the sum on the left - hand side of ( [ sumfor ] ) coincides with the corresponding sum in the vacuum energy , if we take@xmath64 the application of the summation formula ( [ sumfor ] ) with ( [ finapf ] ) allows us to write the vacuum energy from ( [ elr ] ) in the decomposed form@xmath60}=ae_{r^{(d_{1},1)}\times \sigma } + \sum_{j=1,2}e_{j}+\delta e_{[a_{1},a_{2 } ] } , \label{edec}\]]where we have introduced the notations@xmath65and @xmath66}=-\frac{1}{\pi } \int \frac{d\mathbf{k}}{(2\pi ) ^{d_{1}-1}}\sum_{\beta } \int_{\sqrt{k^{2}+m_{\beta } ^{2}}}^{\infty } dz\frac{% \sqrt{z^{2}-k^{2}-m_{\beta } ^{2}}}{\frac{(\beta _ { 1}z-1)(\beta _ { 2}z-1)}{% ( \beta _ { 1}z+1)(\beta _ { 2}z+1)}e^{2az}-1}\left ( a+\sum_{j=1}^{2}\frac{\beta _ { j}}{\beta _ { j}^{2}z^{2}-1}\right ) .
\label{dele0}\]]in eq .
( [ edec ] ) , @xmath67is the vacuum energy ( per unit volume along the directions @xmath68 ) in the spacetime of topology @xmath13 for the case when the plates are absent . in the limit @xmath69 the term @xmath70}$ ] vanishes and the contribution @xmath71 can be interpreted as the vacuum energy ( per unit volume along the directions @xmath72 ) induced by the presence of the plate located at @xmath34 in the half - space @xmath73 .
these single plate components do not depend on the location of the plate and do not contribute to the vacuum force acting on the plates .
as it will be shown below , the latter is determined by the term @xmath74}$ ] .
note that this contribution is finite and that the cutoff function is strictly necessary for the terms @xmath75 and @xmath71 only . in the discussion below we will refer to @xmath74}$ ] as the interaction term . for further simplification of the corresponding expression ,
we use the relation@xmath76 in order to derive this formula , we must first integrate the left - hand side over the angular part of the vector @xmath77 and then change to a new integration variable , @xmath78 . after introducing polar coordinates in the @xmath79-plane and integrating over the polar angle , we get eq .
( [ rel3 ] ) . by using this relation , for the interaction part of the vacuum energy we find @xmath66}=-\frac{(4\pi )
^{-d_{1}/2}}{\gamma ( d_{1}/2 + 1)}% \sum_{\beta } \int_{m_{\beta } } ^{\infty } dx\,\frac{(x^{2}-m_{\beta } ^{2})\,^{d_{1}/2}}{\frac{(\beta _ { 1}x-1)(\beta _ { 2}x-1)}{(\beta _ { 1}x+1)(\beta _ { 2}x+1)}e^{2ax}-1}\left ( a+\sum_{j=1}^{2}\frac{\beta _ { j}}{% \beta _ { j}^{2}x^{2}-1}\right ) .
\label{dele2}\]]using @xmath80 , \label{rel2}\]]and integrating by parts , the interaction term in the vacuum energy can be written as @xmath81 } & = & \frac{(4\pi ) ^{-d_{1}/2}}{\gamma ( d_{1}/2)}% \sum_{\beta } \int_{m_{\beta } } ^{\infty } dx\,x(x^{2}-m_{\beta } ^{2})\,^{d_{1}/2 - 1 } \notag \\ & & \times \ln \left [ 1-\frac{(\beta _ { 1}x+1)(\beta _ { 2}x+1)}{(\beta _ { 1}x-1)(\beta _ { 2}x-1)}e^{-2ax}\right ] \ , .
\label{dele}\end{aligned}\]]in the case when the internal space is absent and for a massless scalar field this result reduces to the one derived in @xcite .
note that the bulk divergences in the vacuum energy between the plates are contained in the first term on the right - hand side of ( [ edec ] ) and the boundary divergences are contained in the single plate contributions @xmath71 .
the interaction part is unambiguously defined . in particular , it does not depend on the regularization scheme used ( see , for example , ref .
@xcite for the case without the internal space , where exactly the same result is obtained with zeta function techniques ) . for the special cases of dirichlet and neumann boundary conditions on both plates , from ( [ dele ] ) one finds@xmath66}^{\mathrm{(j , j)}}=-\frac{(4\pi ) ^{-d_{1}/2}a}{\gamma ( d_{1}/2 + 1)}\sum_{\beta } \int_{m_{\beta } } ^{\infty } dx\,\frac{% ( x^{2}-m_{\beta } ^{2})\,^{d_{1}/2}}{e^{2ax}-1 } , \label{deledn}\]]with @xmath82 and @xmath83 for dirichlet and neumann boundary conditions , respectively . by expanding the factor @xmath84 in the integrand one gets @xmath85being @xmath86 the mac - donald ( or modified bessel ) function .
this allows us to write the corresponding vacuum energy for dirichlet and neumann scalars as @xmath66}^{\mathrm{(j , j)}}=-\frac{2a^{-d_{1}}}{(8\pi ) ^{(d_{1}+1)/2}}\sum_{\beta } \sum_{n=1}^{\infty } \frac{% f_{(d_{1}+1)/2}(2nam_{\beta } ) } { n^{d_{1}+1 } } , \label{deledn1}\]]with the notation@xmath87the energy given by eq . ( [ deledn1 ] )
is always negative and the corresponding casimir forces are attractive for all interplate distances , as will be shown below . in the case
@xmath88 and for a massless scalar field , eq . ( [ deledn1 ] ) reduces to the expression given in ref .
full09b , where the zeta function method was used . for dirichlet boundary conditions on one plate and neumann boundary conditions on the other , similarly to ( [ dele2 ] )
we get@xmath81}^{\mathrm{(d , n ) } } & = & \frac{(4\pi ) ^{-d_{1}/2}a}{% \gamma ( d_{1}/2 + 1)}\sum_{\beta } \int_{m_{\beta } } ^{\infty } dx\,\frac{% ( x^{2}-m_{\beta } ^{2})\,^{d_{1}/2}}{e^{2ax}+1 } \notag \\ & = & -\frac{2a^{-d_{1}}}{(8\pi ) ^{(d_{1}+1)/2}}\sum_{\beta } \sum_{n=1}^{\infty } \frac{f_{(d_{1}+1)/2}(2nam_{\beta } ) } { ( -1)^{n}n^{d_{1}+1}}. \label{deled+n}\end{aligned}\]]in this case the energy @xmath74}$ ] is always positive and the corresponding vacuum forces are repulsive for all distances between the plates . by using the result ( [ dele ] ) , in a similar way as in @xcite , we obtain the corresponding casimir energy for a conformally coupled massless scalar field @xmath89 on the background of a spacetime with metric tensor @xmath90 , where the metric @xmath91 is defined by the line element ( [ metric ] ) .
we assume that the field obeys the boundary conditions : @xmath92% \overline{\varphi } ( x)=0,\;\omega _ { j}=\omega ( x_{j}^{d_{1 } } ) , \label{bcbrane}\]]on two codimension - one branes with coordinates @xmath93,@xmath94 .
the corresponding results for the interaction part of the casimir energy can be derived from those obtained before simply by using the conformal relation that relates the two problems .
the fields are connected by the formula @xmath95 . making use of this relation , from eqs .
( [ boundcond ] ) and ( [ bcbrane ] ) we obtain the following relations between the robin coefficients:@xmath96 ^{-1}\overline{\beta } _ { j } , \label{relbeta}\]]where @xmath97 .
we conclude that for a conformally coupled massless scalar field with boundary conditions ( [ bcbrane ] ) , the interaction part of the vacuum energy in the region between the branes is given by ( [ dele ] ) , where the coefficients @xmath30 are defined by the relations ( [ relbeta ] ) . in particular , for the case of neumann boundary conditions ( @xmath98 ) , one has @xmath99 $ ] . in the special case of the ads bulk used in the randall - sundrum braneworld model @xcite
( note that in this model only the inside region between the branes is considered ) we have @xmath100 , being @xmath101 the ads radius .
the corresponding robin coefficients for an untwisted scalar are given by the relations @xcite
@xmath102where @xmath103 and @xmath104 are the mass parameters in the surface action of the scalar field for the left and right branes , respectively . for a twisted scalar field ,
dirichlet boundary conditions are obtained on both branes .
to summarize , as we see , in the case of the warped geometry the corresponding vacuum energy is not , in general , a monotonic function of the inter - brane distance and can display a minimum , corresponding to the stable equilibrium point .
this property can be used in braneworld models for the stabilization of the radion field .
an important difference between the warped geometry and the one discussed before is that now the single brane contributions to the vacuum energy depend on the location of the brane and , hence , give additional contributions to the force acting on the brane .
the divergences in the single brane components are absorbed by adding the respective counterterms to the brane action .
the coefficients of these counterterms are not computable within the framework of the low - energy effective theory and should be considered as parameters which are fixed by imposing renormalization conditions on the corresponding effective potential ( see also the discussions in refs .
gold00,flac01,garr01,flac03,flac03b , saha06b , pont01 ) .
the vacuum energy corresponding to the region @xmath105 , @xmath106 , @xmath107 will be denoted @xmath108}c_{1}\cdots c_{d_{1}-1}$ ] .
the volume of this region is @xmath109 , being @xmath110 the volume of the internal space .
the corresponding vacuum stress at @xmath111 is given by@xmath112}c_{1}\cdots c_{d_{1}-1}=p_{0}+\delta p(a_{1}+ ) , \label{p}\]]where@xmath113}. \label{p0}\]]using eq . ( [ dele ] ) , we find @xmath114as can be easily seen , the vacuum stress at @xmath115 is given by the same expression : @xmath116 . for the geometry of two parallel plates ,
the total vacuum energy is the sum of the contributions from the regions @xmath117 , @xmath118 and @xmath119 .
when investigating the resulting force on the plate at @xmath120 , in order to deal with finite regions from both sides , we will consider a piston like geometry ( with large transverse dimensions , for a piston with finite cross section see below ) , assuming the presence of an additional plate located at @xmath121 .
for the corresponding vacuum stress at @xmath122 , one has@xmath123 } , \label{p1-}\]]with @xmath124 .
the resulting pressure on the plate at @xmath125 is given by the difference@xmath126as we see , the contributions to the vacuum force coming from the term @xmath127 are the same from the left and from the right sides of the plate , so that there is no netto contribution to the effective force . in the limit @xmath128
one has @xmath129 and the casimir force acting on the plate at @xmath34 , @xmath39 , in the original two - plate geometry is given by the expression @xmath130this force is attractive when @xmath131 and repulsive when @xmath132 . if one does not take into account the contributions from the exterior regions @xmath133 and @xmath134 , the effective pressure is given by ( [ p ] ) where the renormalized value for @xmath127 does not depend on the separation of the plates . in the case
@xmath135 and for periodic boundary conditions along the compactified dimension , the renormalized value @xmath127 is positive ( see next section , eq .
( [ el2 ] ) ) which would correspond to the repulsive force between the plates observed in the first paper of chen06 ( see also the discussion in ref .
@xcite ) . as is clearly seen from eq .
( [ pcas ] ) , the sign of the vacuum stress on the plate is determined by the sign of the integral in this formula . in fig .
[ fig1 ] we have plotted the location of the zeros for this integral on the @xmath136-plane in the case @xmath88 and for different values of @xmath137 ( figures on the curves ) . for a given @xmath138 ,
these zeros are located on two curves symmetric with respect to the line @xmath139 .
the integral is positive in the region containing this line and it is negative outside . in particular , the casimir force between the plates is always attractive for symmetric boundary conditions with @xmath140 .
this result is a special case of the general theorem @xcite , which dictates an attraction between bodies with the _ same _ properties .
note that the curves in fig .
[ fig1 ] display the locations of the zeros for the casimir force in the geometry of two parallel plates on the background of a 4-dimensional minkowski spacetime .
for the special cases of dirichlet and neumann boundary conditions , making use of the recurrence relations for the function @xmath86 , the casimir forces can be written as @xmath141 , \label{fdn}\end{aligned}\]]with @xmath142 .
again , in the case @xmath88 and for a massless scalar field this result reduces to the one obtained in full09b .
the forces described by eq .
( [ fdn ] ) are attractive for all distances between the plates , irrespective of the geometry of the internal subspace . for dirichlet
boundary conditions on one plate and neumann boundary conditions on the other , the expression for the casimir force is obtained from eq .
( [ fdn ] ) after introducing an additional factor @xmath143 in the summation over @xmath61 . in this case ,
casimir forces are repulsive . in the general case of robin boundary conditions
the casimir forces can be either attractive or repulsive , depending on the coefficients present in the definition of the boundary conditions , and on the distance between the plates . for the special case of the topology @xmath144 this issue
will be illustrated in the next section .
let us now consider the asymptotic behavior of the casimir force as a function of the size of the internal space .
note that if the size of the internal space is of the order @xmath145 , then for nonzero modes one has @xmath146 .
for small values of @xmath145 and for the nonzero modes , @xmath147 is large .
the contribution of these modes is exponentially suppressed and the main contribution comes form the zero mode . in this case , from ( [ p1 ] ) we recover the casimir force for two parallel plates in @xmath148-dimensional minkowskian spacetime , namely@xmath149for the case of a degenerated zero eigenstate the corresponding degeneracy factor must be included on the right - hand side . in some models of compactification
the zero mode is absent ( for example , in models with twisted boundary conditions along the compactified dimensions , see below ) .
in such cases , for small values of @xmath145 the main contribution to the casimir force comes from the lowest mode @xmath150 and , to leading order , one gets@xmath151where @xmath152 .
hence , here the casimir forces are exponentially suppressed for small sizes of the internal space . for small values of the inter - plate distance , @xmath153 , the main contribution into the integral in eq .
( [ p1 ] ) comes from larger values of @xmath154 and , to leading order , one has@xmath155except for the case of dirichlet boundary conditions on one plate and non - dirichlet boundary conditions on the other .
we see that in this limit ( [ forcesm ] ) the casimir force is attractive .
however , in the case of dirichlet boundary condition on one plate and non - dirichlet boundary condition on the other the casimir force is repulsive at small distances , what is indeed a remarkable result .
as a simple example of a particular application of the general results obtained above , we will first consider the special case where @xmath135 , with the size of the internal space being @xmath156 .
for the compact dimension we assume a general periodicity condition of the form@xmath157with constant @xmath36 , @xmath158 . the specific cases @xmath159 and @xmath160 correspond to untwisted and to twisted fields , respectively .
the corresponding part of the scalar eigenfunctions is@xmath161the formulas for the topology @xmath144 are obtained from the results given in the previous sections , by taking@xmath162 in particular , for the casimir force one has@xmath163 for large values of @xmath145 the main contribution to the series in ( ps1 ) comes from large values of @xmath164 and one can replace the summation over @xmath164 by an integration .
after some transformations , to leading order we find the result@xmath165which is , in fact , the casimir force for two parallel plates in @xmath0-dimensional minkowskian spacetime .
as we already noted before , depending on the values of the coefficients @xmath30 and of the distance between the plates , the casimir force ( [ ps1 ] ) can be either attractive or repulsive . in fig .
[ fig2 ] , and corresponding to the model with @xmath166 and for a massless scalar field with dirichlet boundary conditions , we have plotted the ratio @xmath167 as a function of @xmath168 , where @xmath169 is the standard casimir force . the values on each of the curves correspond to those of the parameter @xmath170 .
as we have explained before , for @xmath171 the zero mode is absent and for large values of @xmath168 the casimir force is exponentially suppressed . in fig .
[ fig3 ] the casimir force is plotted for the topology @xmath172 and an untwisted ( @xmath173 ) massless scalar field with robin coefficients
@xmath174 , @xmath175 ( @xmath176 a fixed length scale ) as a function of @xmath177 for @xmath178 ( full curve ) and @xmath179 ( dashed curve ) .
the thick curve corresponds to the casimir force for two parallel plates in minkowski spacetime @xmath180 with the same robin coefficients .
as is seen , the corresponding casimir forces are attractive for small and large distances between the plates while they are repulsive for intermediate distances .
there are two equilibrium points corresponding to the zeros of the function @xmath181 .
the leftmost point is unstable whereas the rightmost one is stable .
hence , in this case the casimir force
_ stabilizes _
the distance between the plates . this feature can be used in braneworld models for the stabilization of the radion field .
we see from fig .
[ fig3 ] that the height of the barrier between the stable and the unstable equilibrium points is increased by the presence of the internal space . as a consequence ,
an enhancement of the repulsive casimir effect , coming from the extra dimension , occurs .
as already explained before , in the case of dirichlet boundary conditions on one plate and non - dirichlet boundary conditions on the other , the casimir force is repulsive at small separations . in fig .
[ fig4 ] we illustrate this feature for the topology @xmath182 in the case of an untwisted massless scalar field with @xmath183 , @xmath175 . as in fig .
[ fig3 ] , the full ( dashed ) curve stands for @xmath184 ( @xmath179 ) and the thick curve corresponds to the casimir force for two parallel plates in minkowski spacetime @xmath180 with the same robin coefficients . for the topology under consideration , the casimir energy for the bulk without boundaries,@xmath185can be further simplified through the abel - plana summation formula , in the form @xcite@xmath186what leads to the result@xmath187the second term on the right - hand side of this expression is finite and introduction of a cutoff function is necessary for the first term only .
note that the latter is the vacuum energy density for the spatial topology @xmath188 and , hence , the second term on the right - hand side of eq .
( [ el2 ] ) is the contribution to the vacuum energy induced by the compactness of the @xmath189 dimension . in particular ,
the topological part of the vacuum energy is always negative ( positive ) for untwisted ( twisted ) scalars . in a similar way , we can also extract the topological contributions in the single plate terms of the vacuum energy .
after applying the summation formula ( [ sumform1 ] ) and after integration , these contributions yield@xmath190where @xmath191is the vacuum energy ( per unit volume along the coordinates @xmath192 ) for a single plate in minkowski spacetime with trivial topology @xmath193 .
hence , the last two terms on the right - hand side of ( [ ejs1 ] ) are the terms in the vacuum energy corresponding to a single plate and due to the compactness of the dimension @xmath189 .
these terms are finite and renormalization is needed for the minowskian part @xmath194 only . note that for dirichlet and neumann boundary conditions the last term on the right of eq .
( [ ejs1 ] ) vanishes .
the case of a @xmath16-dimensional torus as internal space , @xmath195 , can be considered in a similar way . for a scalar field with the periodicity condition
@xmath196 along the coordinate @xmath197 , @xmath198 , the formulas for the casimir energy and force are obtained from the general expressions in sects .
sec : casen and [ sec : casforce ] with the substitutions@xmath199 concerning the issues of embedding the model in string theory and of the discussions of the holographic principle there , the case of the internal space @xmath200 is of very special interest .
the corresponding eigenfunctions @xmath41 are expressed in terms of spherical harmonics of degree @xmath201 , @xmath202 .
for the internal space with radius @xmath145 the expressions for the casimir energy and casimir force are quite easily obtained from the general formulas given in the above sections , just by replacing@xmath203here the factor under the summation sign is the degeneracy of the angular mode with a given @xmath201 .
in a way very much similar to the procedure described in the preceding sections , we are able to treat the more general case when a part of the dimensions @xmath59 are still constrained by boundary conditions .
this corresponds to considering a generalized piston geometry , a quite fashionable situation nowadays , in particular , in the quest for negative casimir forces ( for the investigation of the casimir effect in a piston geometry see @xcite and references therein ) .
here we have obtained those in the configurations above , but would like to see now the differences introduced in our results by the consideration of piston geometries .
we will denote by @xmath204 the number of unconstrained dimensions ( coordinates @xmath205 ) and by @xmath206 , with a collective index @xmath207 , the eigenvalues of the laplacian along the constrained directions:@xmath208the eigenfrequencies in the region between the plates are here given by @xmath209with the vacuum energy being@xmath60}=\frac{1}{2}\int \frac{d\mathbf{k}_{d_{1}}}{(2\pi ) ^{d_{1}}}% \sum_{i,\beta } \sum_{n=1}^{\infty } \sqrt{k_{d_{1}}^{2}+z_{n}^{2}/a^{2}+% \gamma _ { i}^{2}+m_{\beta } ^{2}}. \label{epist}\]]after applying the summation formula ( [ sumfor ] ) to the sum over @xmath61 , we can write the energy in the decomposed form@xmath60}=\frac{a}{2}\int \frac{d\mathbf{k}_{d_{1}+1}}{(2\pi ) ^{d_{1}+1}}\sum_{i,\beta } \sqrt{k_{d_{1}+1}^{2}+\gamma _ { i}^{2}+m_{\beta } ^{2}}+\sum_{j=1,2}e_{j}+\delta e_{[a_{1},a_{2}]}. \label{edecpist}\]]here the expressions for the terms @xmath71 and @xmath74}$ ] are obtained from eqs .
( [ ej ] ) and ( [ dele0 ] ) by the replacements@xmath210these formulas are further simplified after integrating over the angular part of the vector @xmath211 .
the corresponding expressions are obtained from the results of sects .
[ sec : casen ] and [ sec : casforce ] , with the replacements ( [ replace2 ] ) .
in particular , for the interaction part of the casimir energy ( per unit volume along the direction @xmath212 ) one has@xmath81 } & = & \frac{(4\pi ) ^{-(d_{1}+1)/2}}{\gamma ( ( d_{1}+1)/2)}\sum_{i,\beta } \int_{\sqrt{\gamma _ { i}^{2}+m_{\beta } ^{2}}% } ^{\infty } dx\,x(x^{2}-\gamma _ { i}^{2}-m_{\beta } ^{2})\,^{(d_{1}-1)/2 } \notag \\ & & \times \ln \left [ 1-\frac{(\beta _ { 1}x+1)(\beta _ { 2}x+1)}{(\beta _ { 1}x-1)(\beta _ { 2}x-1)}e^{-2ax}\right ] \ , .
\label{delepist}\end{aligned}\]]the expression for the casimir pressure takes the form @xmath213where @xmath214 is the volume of the piston cross section along the coordinates @xmath215 .
in particular , for dirichlet and neumann boundary conditions on the plates , we find@xmath216 _ { z=2na\sqrt{\gamma _ { i}^{2}+m_{\beta } ^{2 } } } , \label{pjjpist}\]]@xmath142 , the function @xmath217 being defined in eq .
( [ fnu ] ) .
the corresponding force remains attractive independently of the form of the cross section . in the case of dirichlet boundary condition on one plate and neumann boundary condition on the other the expression for the casimir force
is obtained from ( [ pjjpist ] ) after introducing an additional factor @xmath218 , and the resulting force is always repulsive . in the special case @xmath219 and for a massless scalar field , eq .
( [ pjjpist ] ) reduces to the formula for the casimir force in ref .
@xcite , as it should . on the base of eqs .
( [ delepist ] ) and ( [ pcaspist ] ) we can analyze the geometry of a generalized piston with two chambers , assuming that the plates are located at @xmath220 , with the robin coefficient @xmath221 for the left plate ( for the casimir effect in a piston geometry see , for example , refs .
@xcite ) . for an arbitrary cross section ,
the effective pressure on the plate at @xmath120 is given by @xmath222with @xmath124 .
for @xmath223 ( @xmath224 ) the resulting force on the plate is directed towards the right ( left ) plate . for dirichlet and neumann boundary conditions
the casimir stress given by eq .
( [ pjjpist ] ) is a monotonic function of the plate separation and , hence , in the piston geometry with two chambers the resulting force ( [ pa1pist ] ) is directed toward the closer plate . in the special case of the geometry of a piston with circular cross section in the plane @xmath225 ,
the corresponding part in the eigenfunctions has the form @xmath226 , @xmath227 , where @xmath228 and @xmath229 are polar coordinates on this plane and @xmath230 is the bessel function .
the eigenvalues for the quantum number @xmath231 are quantized by the boundary conditions on the cylindrical surface @xmath232 , with @xmath233 being the piston radius .
for example , in the case of a dirichlet boundary condition one has @xmath234 , where @xmath235 is the @xmath236-th positive zero of the function @xmath237 .
the corresponding formulas for the casimir energy and forces are obtained from the general results ( [ delepist ] ) and ( [ pcas ] ) with the substitutions@xmath238other types of boundary conditions on the cylindrical boundary can be considered in a similar way . moreover ,
the generalization of our procedure to more than three plates is also easy to carry out .
we have investigated in this paper the influence of extra dimensions on the casimir energy and on the casimir force for a massive scalar field with an arbitrary curvature coupling parameter , in the usual geometry of two parallel plates .
we have assumed that on the plates the field obeys robin boundary conditions with , in general , different coefficients for the two different plates .
the corresponding eigenfrequencies are expressed in terms of solutions of a transcendental equation ( [ eigeq ] ) , thus they are known implicitly only . by applying the summation formula ( [ sumfor ] ) to the corresponding series in the mode - sum for the vacuum energy in the region between the plates , we have explicitly extracted , in a cut - off independent way , the boundary - free ( topological ) part and the contributions induced by the single plates ( when the other plate is absent ) .
the remaining interaction part is finite for all nonzero inter - plate distances and is cut - off independent .
the surface divergences in the casimir energy are contained in the single plate components only .
but the latter do not depend on the location of the plate and do not contribute to the casimir force .
for an arbitrary internal space , the interaction part of the casimir energy is given by eq .
( [ dele ] ) . in the special cases of dirichlet and neumann boundary conditions on both plates this formula leads to the result ( [ deledn1 ] ) and
the corresponding energy is always negative .
for dirichlet boundary conditions on one plate and neumann boundary conditions on the other , the interaction component of the vacuum energy is given by eq .
( [ deled+n ] ) , and it is positive for all values of the interplate distance . in the case of a conformally coupled massless field on the background of a spacetime conformally related to the one described by the line element ( [ metric ] ) with the conformal factor @xmath239 , the interaction part of the casimir energy
is given by eq .
( [ dele ] ) , with the coefficients @xmath240 being related to the specific coefficients of the robin boundary conditions ( [ bcbrane ] ) and to the conformal factor by eqs .
( [ relbeta ] ) . in the randall - sundrum two brane model with a compact internal space ,
the corresponding robin coefficients are given by eq .
( [ betajrs ] ) and the corresponding vacuum energy can have a minimum , corresponding to the stable equilibrium point .
this feature is useful in braneworld models for the stabilization of the radion field .
the interaction forces between the plates for the most general case of internal space have been considered in sect .
[ sec : casforce ] . in order to obtain the resulting force , the contributions from both sides of the plates must be taken into account .
then , the forces coming from the topological parts of the vacuum energy cancel out and only the interaction terms contribute to the casimir force . in order to show this important fact explicitly ,
we have considered a piston - like geometry , by introducing a third plate . at the end of the calculation this plate
is sent to infinity .
the resulting casimir force is given by eq .
( [ pcas ] ) . with independence of the geometry of the internal space ,
the force is attractive for dirichlet or neumann boundary conditions on both plates ( formula ( [ fdn ] ) ) and it is repulsive for dirichlet boundary conditions on one plate and neumann boundary conditions on the other . in both cases
the force is a monotonic function of the distance . for general robin boundary conditions
the casimir force can be either attractive ( corresponding to negative values of @xmath181 ) or repulsive ( positive values of @xmath181 ) , depending on the particular robin coefficients and on the distance between the plates . for small values of the size of the internal space and in models where the zero modes along the internal space are present
, the main contribution to the casimir force comes from the zero modes and the contributions of the nonzero modes are exponentially suppressed . in this limit , to leading order we recover the standard result for the casimir force between two plates in @xmath15-dimensional minkowski spacetime .
when the zero mode is absent ( for example , in the case of twisted boundary conditions along the compactified dimensions ) , the casimir forces are exponentially suppressed in the limit of small size of the internal space . for small values of the inter - plate distance
the casimir forces are attractive , independently of the values of the robin coefficients , except for the case of dirichlet boundary conditions on one plate and non - dirichlet boundary conditions on the other . in this latter case ,
the casimir force is repulsive at small distances .
it is interesting to remark that this property could be used in the proposal of a casimir experiment with the purpose to carry out an explicit detailed observation of ` large ' extra dimensions as allowed by some models of particle physics . as an illustration of the general results , in sect .
[ sec : example ] we have considered a special model for the internal space @xmath135 , with the periodicity condition ( [ percond ] ) along the compactified dimension . for the specific values @xmath241 this condition corresponds to untwisted and twisted scalar fields , respectively . in fig .
[ fig2 ] , for dirichlet boundary conditions we depicted the dependence of the casimir force on the distance between the plates for different values of the parameter @xmath170 . in fig .
[ fig3 ] a plot of the casimir force in the case of robin boundary conditions on both plates as a function of the inter - plate distance has been provided . in the example
considered , the casimir force is attractive both for large and for small distances , while it is repulsive at intermediate distances .
the casimir force vanishes at two values of the inter - plate distance , which correspond to equilibrium points .
the leftmost point is unstable and the rightmost one is locally stable .
as shown in the plot , the stability of the rightmost equilibrium point is enhanced by the presence of the internal space .
formulas for the casimir energy and force for the more general internal spaces @xmath242 ( @xmath16-torus ) and @xmath243 have been obtained from the general results of sects .
[ sec : casen ] and sec : casforce . in the last section [ sec : piston ] , we have extended the results from the previous ones to the case of a piston geometry with finite cross section of arbitrary form along some subset of the dimensions .
the corresponding expressions for the casimir energy and casimir force between the plates are given by eqs .
( [ delepist ] ) and ( [ pcaspist ] ) .
we have checked that the qualitative features described above remain basically unaltered . in particular , the possibility of a repulsive casimir effect is again observed . in the special case of a piston geometry of circular cross section on the plane @xmath225 ,
the corresponding formulas have been specified in ( [ replace3 ] ) .
the search for specific applications of this study to practical situations in braneworld models , nano - physics and particle physics will keep us busy for some time .
e.e . has been supported by miciin ( spain ) , project fis2006 - 02842 , and by agaur ( generalitat de catalunya ) , contract 2005sgr-00790 and grant dgr2008be1 - 00180 .
s.d.o . was supported in part by projects fis2006 - 02842 and pie2007 - 50i023 ( miciin , spain ) .
a.a.s . was supported by the esf programme
new trends and applications of the casimir effect " and in part by the armenian ministry of education and science grant no . 119 .
a.a.s . gratefully acknowledges the hospitality of the abdus salam international centre for theoretical physics ( trieste , italy ) where part of this work was done .
mostepanenko and n.n .
trunov , _ the casimir effect and its applications _ ( clarendon , oxford , 1997 ) ; m. bordag , u. mohidden , and v.m .
mostepanenko , phys . rep . * 353 * , 1 ( 2001 ) ; k.a .
milton , _ the casimir effect : physical manifestation of zero - point energy _
( world scientific , singapore , 2002 ) ; e. elizalde , s.d .
odintsov , a. romeo , a.a .
bytsenko and s. zerbini , _ zeta regularization techniques with applications _ ( world scientific , singapore , 1994 ) ; e. elizalde , _ ten physical applications of spectral zeta functions _ , lecture notes in physics ( springer - verlag , berlin , 1995 ) ; m.j .
duff , b.e.w .
nilsson , and c.n .
pope , phys . rep .
* 130 * , 1 ( 1986 ) ; a.a .
bytsenko , g. cognola , l. vanzo , and s. zerbini , phys . rep . * 266 * , 1 ( 1996 ) .
milton , grav . cosmol . * 9 * , 66 ( 2003 ) ; e. elizalde , j. phys .
a * 39 * , 6299 ( 2006 ) ; b. green and j. levin , j. high energy phys .
* 11 * , 096 ( 2007 ) ; p.
burikham , a. chatrabhuti , p. patcharamaneepakorn , and k. pimsamarn , j. high energy phys . * 07 * , 013 ( 2008 ) .
mostepanenko and i.yu .
sokolov , phys .
a * 125 * , 405 ( 1987 ) ; j. c. long , h. w. chan and j. c. price , nucl .
b * 539 * , 23 ( 1999 ) ; r.s .
decca , d. lpez , e. fischbach , g.l .
klimchitskaya , d.e .
krause and v.m .
mostepanenko , ann . phys . * 318 * , 37 ( 2005 ) ; phys .
d * 75 * , 077101 ( 2007 ) .
cavalcanti , phys .
d * 69 * , 065015 ( 2004 ) ; m.p .
hertzberg , r.l .
jaffe , m. kardar , and a. scardicchio , phys .
lett . * 95 * , 250402 ( 2005 ) ; g. barton , phys . rev .
d * 73 * , 065018 ( 2006 ) ; v.n .
marachevsky , phys .
d * 75 * , 085019 ( 2007 ) ; a. edery , phys . rev .
d * 75 * , 105012 ( 2007 ) ; x. zhai and x. li , phys .
d * 76 * , 047704 ( 2007 ) ; a. edery and v.n .
marachevsky , phys .
d * 78 * , 025021 ( 2008 ) . | we evaluate the casimir energy and force for a massive scalar field with general curvature coupling parameter , subject to robin boundary conditions on two codimension - one parallel plates , located on a @xmath0-dimensional background spacetime with an arbitrary internal space . the most general case of different robin coefficients on the two separate plates is considered . with independence of the geometry of the internal space ,
the casimir forces are seen to be attractive for special cases of dirichlet or neumann boundary conditions on both plates and repulsive for dirichlet boundary conditions on one plate and neumann boundary conditions on the other . for robin boundary conditions ,
the casimir forces can be either attractive or repulsive , depending on the robin coefficients and the separation between the plates , what is actually remarkable and useful .
indeed , we demonstrate the existence of an equilibrium point for the interplate distance , which is stabilized due to the casimir force , and show that stability is enhanced by the presence of the extra dimensions
. applications of these properties in braneworld models are discussed . finally , the corresponding results are generalized to the geometry of a piston of arbitrary cross section .
pacs numbers : 03.70.+k , 11.10.kk |
there are general reasons for the interest in the possibility of cp violation in the scalar sector .
cp violation is a necessary ingredient for the generation of the baryon asymmetry of the universe @xcite .
it is believed that cp violation in the kobayashi
maskawa matrix is not large enough to explain that asymmetry @xcite .
it has been speculated @xcite that the scalar sector might provide the missing cp violation .
there have been studies of possible signatures of cp violation in the scalar sector .
it has been remarked @xcite that the simultaneous presence of the three couplings @xmath8 , @xmath9 , and @xmath10 , where @xmath11 , @xmath12 and @xmath13 are three neutral scalar fields in any model , implies cp violation .
similarly , the simultaneous presence of the three couplings @xmath14 , @xmath15 , and @xmath16 , represents cp violation .
this is because the c quantum number of the @xmath17 is @xmath18 .
another work @xcite has considered various cp - violating lagrangians including scalars , fermions , and vector bosons , and has suggested looking for cp violation in the decay mode @xmath19 occurring when , in the rest frame of the decaying neutral scalar @xmath20 , the momentum distribution of the @xmath21 is not the same as the momentum distribution of the @xmath22 or , in a similar fashion , in @xmath23 .
the first of these cp - violating asymmetries has later been computed @xcite in the context of the two - higgs - doublet model .
however , the decay mode @xmath19 is phase - space disfavored as compared to the simpler decay modes @xmath24 and @xmath25 .
other studies @xcite have concentrated on cp - violating phenomena originating in the interplay of scalars and fermions , in particular the effects of top - quark physics .
the aim of this work is the computation of a cp - violating asymmetry in the two - higgs - doublet model without any fermions .
the model has gauge symmetry su(2)@xmath26u(1 ) , which is spontaneously broken to the u(1 ) of electromagnetism by the vacuum expectation values ( vev s ) of the two higgs doublets .
i look for cp violation involving solely the gauge interactions of the scalars . for simplicity ,
i do not consider the presence of fermions , which presence would lead to extra sources of cp violation , both in the fermion sector , and in the interplay of the fermion and the scalar sectors .
i also omit possible sources of cp violation in the cubic and quartic interactions of the physical scalars . those scalars are two charged particles @xmath27 , with mass @xmath28 , and three neutral particles @xmath1 , @xmath2 and @xmath3 , with masses @xmath29 ,
@xmath30 and @xmath31 , respectively
. besides , the spectrum of the model includes the massive intermediate vector bosons @xmath32 and @xmath17 , with masses @xmath33 gev and @xmath34 gev respectively , and the massless photon . for a fairly large range of the masses of the scalars , either the two decays @xmath4 and @xmath5 , or the two decays @xmath6 and @xmath7 ,
are kinematically allowed ( the neutral scalars may be numbered so that @xmath1 is the scalar for which one of these couples of decays is allowed ) . then ,
the possibility of a cp - violating difference between the rate of one decay and the rate of its cp - conjugated decay exists .
it is my purpose to calculate that difference .
it has recently been observed @xcite that the two - higgs - doublet model has one and only one source of cp violation in the gauge interactions of the scalars .
i describe it briefly .
because the u(1 ) of electromagnetism is preserved in the symmetry breaking , we can , without loss of generality , choose a basis for the two scalar doublets in which only one of them , @xmath35 , has a vev @xmath36 , while the second one , @xmath37 , does not have a vev .
the two doublets in that basis can be written @xmath38 @xmath39 and @xmath40 are the goldstone bosons , which become the longitudinal components of the @xmath21 and @xmath17 , respectively .
@xmath41 , @xmath42 , and @xmath43 are linear combinations of the three neutral scalar fields @xmath1 , @xmath2 , and @xmath3 , which are the eigenstates of mass . those linear combinations are given by an orthogonal matrix @xmath44 , @xmath45 without loss of generality , we can assume @xmath44 to have determinant @xmath46 .
then , the following useful identities follow : @xmath47 where @xmath48 is the totally antisymmetric tensor with @xmath49 .
there is cp violation in the gauge interactions of the scalars @xcite if and only if @xmath29 , @xmath30 and @xmath31 are all different , and @xmath50 is non - zero .
the quantity @xmath0 has in the two - higgs - doublet model a role analogous to the one of jarlskog s
@xcite @xmath51 in the three - generation standard model .
notice however that , here , there are in principle other sources of cp violation , in the cubic and quartic interactions of the scalars .
i will neglect those extra sources of cp violation throughout this work .
it is important to remark that , though @xmath0 represents cp violation in the mixing of the three neutral scalars , this source of cp violation has nothing to do with the fermions and with the identification of , say , @xmath41 and @xmath42 as being scalars , and @xmath43 as being a pseudoscalar .
that identification can only be done when a specific yukawa lagrangian , coupling the two scalar doublets to the fermion sector , is considered , which i do not do here . specifically , as is clear from eq .
[ j1 ] , @xmath42 and @xmath43 play a completely equivalent role in @xmath0 indeed , as long as there are no yukawa couplings , @xmath42 and @xmath43 may rotate into each other by a simple u(1 ) rephasing of @xmath37 . also , @xmath0 can not be the source of , say , cp violation in the kaon system .
if fermions are introduced in the model , the mixing of the neutral scalars will in principle lead to more cp violation than simply @xmath0 , because of the yukawa interactions of the scalars with the fermions .
let us consider how cp violation proportional to @xmath0 arises in the decay modes that i consider here .
all the vertices needed for the calculations in this paper have been listed in figure 1 .
the tree - level diagram with incoming particles @xmath22 , @xmath52 and @xmath1 is proportional to @xmath53 , while the diagram with incoming particles @xmath21 , @xmath54 and @xmath1 is proportional to @xmath55 . now take a look at figure 2 , in which all the one - loop diagrams which lead to cp violation when interfering with the tree - level diagram are collected . consider for instance the first diagram , with a loop of @xmath56 , and then @xmath2 , as an intermediate state
that diagram is equal to @xmath57 , times a certain momentum integral @xmath58 .
the seven factors of @xmath59 come , three from the vertices , three from the propagators , and one from the wick rotation in the momentum integral .
therefore , the interference of this diagram with the tree - level one is proportional to the real part of @xmath60 .
the @xmath61-matrix factor has an imaginary part equal to @xmath0 .
therefore , if the momentum integral has an absorptive ( _ i.e. _ , imaginary ) part , then the interference term will include @xmath0 times that absorptive part .
this is cp - violating . the absorptive part of the integral plays in the calculation the role of a cp - invariant final - state - interaction phase , which allows @xmath0 to manifest itself .
we can check in a similar fashion that the absorptive parts of all other nine one - loop diagrams in figure 2 lead , when one considers the interference of those diagrams with the tree - level one , to cp violation .
indeed , a careful study of the model and all its vertices shows that the ten diagrams in figure 2 are the only ones which lead to cp violation proportional to @xmath0 in this process .
the cp violation manifests itself in a difference of the decay rates of @xmath62 and @xmath63 , or of @xmath6 and @xmath7 , whichever pair of decays is kinematically allowed .
let me be more explicit . at tree - level ,
the amplitude for the decay @xmath64 , or for the decay @xmath6 , is @xmath65 , from the tree - level vertex . here
, @xmath66 is the polarization vector of the outgoing @xmath21 , and @xmath67 is the incoming momentum of the @xmath27 .
this is because the polarization vector of the @xmath21 is orthogonal to the momentum of that vector boson . at one - loop level , for the same reason , each diagram in figure 2 contributes @xmath68 .
here , @xmath69 are the various @xmath59 factors and @xmath61-matrix factors from the vertices and propagators in the diagram , and @xmath58 is the momentum integral , with the @xmath59 coming from the wick rotation . the amplitudes for the cp - conjugated decays
are , at tree - level , @xmath70 , and , at one - loop level from figure 2 , @xmath71 .
notice that , while the momentum integral is the same , the vertex factors are complex - conjugated .
then , the cp - violating asymmetries are @xmath72 \left [ { \rm or } \hspace{1 mm } \frac{{\rm br } ( h^+ \rightarrow x_1 w^+ ) - { \rm br } ( h^- \rightarrow x_1 w^-)}{{\rm br } ( h^+ \rightarrow x_1 w^+ ) + { \rm br } ( h^- \rightarrow x_1 w^- ) } \right ] & \approx & 2 g^2\ , \sum_{k=1}^{10}\ , \frac{{\rm i m } [ ( t_{21 } + i t_{31 } ) c_k]\ , { \rm re } ( i i_k)}{1 - t_{11}^2}\ , , \label{asymmetry}\end{aligned}\ ] ] where i used the orthogonality of @xmath44 to write @xmath73 . the imaginary part of @xmath74 is simply @xmath0 times a number typically @xmath75 or @xmath76 .
the momentum integral @xmath58 ( the @xmath59 is from the wick rotation ) has a real part if cuts in the corresponding diagram lead to absorptive parts .
notice that in eq .
[ asymmetry ] i have used the approximation of taking , in the denominator , only the square of the modulus of the tree - level contribution to the amplitude .
it is clear from eq .
[ asymmetry ] that the asymmetry will be of the form @xmath77 where @xmath78 represents the sum of the absorptive parts of all the diagrams in figure 2 , weighted by appropriate numbers @xmath75 or @xmath76 ( see the preceding paragraph ) .
as one is interested in how large the asymmetry can be , i now consider the mixing - matrix factor in eq . [ form ] .
because one is constrained by the orthogonality condition @xmath79 , it is obvious that that factor will be maximal when @xmath80 , and we will then have @xmath81 clearly , the order of magnitude of this quantity @xmath82 is 1 .
it is at this point important to remark that , in this specific example , the cp - violating asymmetry approaches its maximum when the decay rate decreases . indeed , as @xmath83 , the asymmetry becomes potentially larger ( as long as @xmath84 remains equal to @xmath85 ) , but the decay rate , which is proportional to @xmath86 , approaches zero .
similarly , the decay rate becomes larger if @xmath87 , but then @xmath88 and the cp asymmetry also vanishes .
this situation is reminiscent of the case of cp - violating asymmetries in decay modes of the @xmath89 mesons , which are generally predicted to be larger when the branching ratios are smaller , and vice - versa .
a way to check that the diagrams in figure 2 are all the relevant ones is to check whether their computation yields a gauge - invariant result . because i just want to compute the absorptive parts of the diagrams , which are finite , it is sufficient to compute the diagrams in the unitary gauge , in which no goldstone bosons and no ghosts are present .
however , in a general t - hooft gauge , in each of the diagrams in figure 2 , a @xmath90 can be used instead of the @xmath32 , or a @xmath40 can be used instead of the @xmath17 . in the two diagrams which have a loop only of @xmath32 or of @xmath17 ( diagrams 1 and 2 ) , ghost loops must also be considered in an t - hooft gauge .
now , in an t - hooft gauge , the @xmath32 propagator contains an extra piece ( relative to the unitary gauge ) in which the @xmath32 has an unphysical squared mass @xmath91 .
similarly , the charged goldstone bosons @xmath90 and the charged ghosts @xmath92 have unphysical squared mass @xmath91 in an t - hooft gauge . when @xmath91 is not infinite as in the unitary gauge , each diagram by itself contains a @xmath93-dependent absorptive part . however , all those unphysical absorptive parts must cancel out when one considers the whole set of diagrams .
the same thing can be said about the propagators of the @xmath17 , which has a piece with unphysical squared mass @xmath94 , and of the goldstone boson @xmath40 and ghost @xmath95 , which have squared mass @xmath94 .
( in principle , @xmath96 . ) the whole set of propagators is given in figure 3 .
the sum over all diagrams of all the absorptive parts must be independent of both @xmath91 and @xmath94 .
i have checked that independence . to be sure
, gauge - independence only applies to an observable quantity .
thus , gauge - independence in this case only occurs when 1 ) the three external particles are all on mass shell , that is , @xmath97 , @xmath98 , and @xmath99 , where @xmath100 , @xmath101 and @xmath102 are the incoming momenta of the external @xmath27 , @xmath32 , and @xmath1 , respectively ; 2 ) one suppresses from the amplitude all terms proportional to @xmath103 , because the amplitude must be multiplied by the polarization vector @xmath66 of the @xmath32 , and @xmath104 ; 3 ) one considers , from each one - loop diagram , only the part which is proportional to @xmath0 upon interference with the tree - level amplitude ; 4 ) one only considers the absorptive part of each one - loop diagram .
one has also to take into consideration that the gauge - dependent absorptive parts sometimes cancel between two similar diagrams with intermediate virtual particle @xmath3 instead of @xmath2 , whenever those absorptive parts do not depend on the mass of that intermediate particle ( @xmath31 or @xmath30 ) ; while other absorptive parts cancel among different - looking diagrams .
i present in this section the results of the computation of each of the ten diagrams in figure 2 .
i am only interested in the absorptive part of each diagram .
those absorptive parts are simpler to compute when the power of the integration momentum @xmath105 in the denominator is lower ; in particular , if only one propagator @xmath106 occurs in the denominator , there is no absorptive part .
moreover , i want to display explicitly how the cancellations which lead to gauge invariance of the whole result occur .
therefore , my method has been the following .
first i have added , for each particular diagram , the contributions with gauge bosons , the ones with goldstone bosons , and the ones with ghosts .
then , i have simplified as much as possible the numerators of the integrands , in such a way that factors in some terms of those numerators cancel out some of the propagators in the denominators , thereby reducing the overall power of the integration momentum @xmath105 in the denominator . as a result
, each diagram becomes the sum of three parts : one piece with an overall @xmath107 in the denominator , leading to no absorptive part ; one piece with @xmath108 in the denominator ; and for the diagrams 7 through 10 only one piece with @xmath109 in the denominator .
i write down here the two latter pieces only , because they are the only ones which lead to absorptive parts in the integrals . for each diagram
, i have taken separately the factors with @xmath61-matrix elements , and all the @xmath59 factors from both the vertices and the propagators , multiplied that by the factor @xmath110 arising in the interference with the tree - level diagram , and taken the imaginary part of the result .
that imaginary part is always a multiple of @xmath0 .
this i did because it is only the quantities @xmath111 $ ] which are relevant for the computation of the asymmetry , according to eq .
[ asymmetry ] . also , it is only for these quantities @xmath111 $ ] that the cancellation of the gauge dependence must occur . as a consequence ,
each momentum integral presented below should be looked upon as being an @xmath58 in the notation of section 2 ( the @xmath59 arising when the wick rotation is performed ) , and only the absorptive parts of each such momentum integral are meaningful in this context .
each diagram in figure 2 is in reality two diagrams , one in which an intermediate state @xmath2 appears , and another one in which the intermediate state is @xmath3 .
those two diagrams have , as is easily seen , opposite signs of @xmath0 in the interference term @xmath111 $ ] .
therefore , if the corresponding momentum integrals happen not to depend on the masses @xmath30 or @xmath31 because the propagator of @xmath2 or of @xmath3 might have been cancelled out by a similar factor in the numerator of the integrand then those terms become irrelevant for the present computation . as a consequence , for each diagram ,
i only write down those integrals which do depend on @xmath30 or @xmath31 . as before ,
@xmath102 , @xmath101 and @xmath100 are the incoming momenta of the scalar @xmath1 , the gauge boson @xmath22 , and the charged scalar @xmath52 .
i have systematically set all three external particles on mass - shell , and used whenever possible @xmath104 in order to simplify the amplitudes .
i first define the various combinations of the integration momentum @xmath105 and of the incoming momenta , and of the masses , which appear in the denominators of the momentum integrals .
they are @xmath112 notice that only @xmath113 depends on @xmath30 .
it is convenient to define the `` triangular function '' @xmath114 : @xmath115 this function is negative if and only if one can form a triangle with sides of length @xmath116 , @xmath117 , and @xmath118 .
i now present the results for the first three diagrams .
diagram 1 : @xmath119 diagram 2 : @xmath120 ( there is an extra factor @xmath121 in diagram 2 relative to diagram 1 , which is a symmetry factor , due to the two identical @xmath122 s in the loop . )
diagram 3 : @xmath123 the results for each of these three diagrams are separately gauge - invariant .
this is because the gauge - dependent piece in each of them ( the piece depending on either @xmath91 or @xmath94 ) is the same , for each case , in the diagram with an intermediate @xmath2 , and in the diagram with an intermediate @xmath3 .
as the sign of the @xmath0 factor is opposite , the gauge - dependent piece cancels out between the diagrams with intermediate @xmath2 and those with intermediate @xmath3 .
the same phenomenon partially occurs in all other diagrams but , in each of them individually , some gauge dependence always remains , as is seen in the following .
diagram 4 : @xmath124 diagram 5 : @xmath125 } \right . \nonumber\\ & + & \left .
\frac{1}{d_4 d_5}\ , \left ( - \frac{p_h^{\nu}}{2}\ , + k^{\nu}\ , \frac{m_z^2 - m_2 ^ 2 + m_1 ^ 2}{4 m_z^2 } \right ) - ( m_2 \rightarrow m_3 ) \right\}\ , .
\label{m5}\end{aligned}\ ] ] diagram 6 : @xmath126 } \right .
\nonumber\\ & + & \left .
\frac{\lambda(m_z^2 , m_1 ^ 2 , m_2 ^ 2)}{d_4 d_5 m_w^2 ( m_1 ^ 2 - m_3 ^ 2)}\ , - ( m_2 \leftrightarrow m_3 ) \right\}\ , .
\label{m6}\end{aligned}\ ] ] diagram 7 : @xmath127 } \right .
\nonumber\\ & + & \left .
\frac{1}{d_4 d_6 } + \frac{m_2 ^ 2 - m_1 ^ 2 - m_z^2}{m_z^2}\ , \frac{1}{d_4 d_5 } + ( 2 m_w^2 + m_z^2 - 2 m_h^2 - m_1 ^ 2 - m_2 ^ 2 ) \frac{1}{d_4 d_5 d_6 } \right . \nonumber\\ & - & \left .
( m_2 \leftrightarrow m_3 ) \right\}\ , .
\label{m7}\end{aligned}\ ] ] diagram 8 : @xmath128\ , .
\label{m8}\end{aligned}\ ] ] diagram 9 : @xmath129}\ , + \frac{(k - p_h)^{\nu}}{4 m_w^2}\ , \frac{(m_h^2 - m_2 ^ 2)}{d_4 ( k^2 - w ) } \right . \nonumber\\ & + & \frac{(k - p_h)^{\nu}}{4}\ , \left ( \frac{1}{d_2 d_4}\ , - \frac{1}{d_1 d_4 } \right ) - \frac{(k - p_h)^{\nu}}{4}\ , \frac{m_h^2 - m_2 ^ 2}{m_w^2}\ , \left ( \frac{1}{d_2 d_4}\ , + \frac{1}{d_1 d_4 } \right ) \nonumber\\ & + & \frac{1}{d_1 d_2 d_4}\ , \left [ \frac{- 6 m_w^2 + 2 m_h^2 + m_1 ^ 2}{4}\ , p_h^{\nu } + \frac{4 m_w^2 - 2 m_2 ^ 2 - m_1 ^ 2}{4}\ , k^{\nu } \right .
\nonumber\\ & + & \left . \left .
\frac{m_1 ^ 2 ( m_h^2 - m_2 ^ 2)}{4 m_w^2}\ , ( k - p_h)^{\nu } \right ] - ( m_2 \leftrightarrow m_3 ) \right\}\ , . \label{m9}\end{aligned}\ ] ] diagram 10 : @xmath130}\ , k^{\nu}\ , \frac{m_2 ^ 2}{m_w^2 } \right . \nonumber\\ & + & \frac{1}{d_4 [ ( k + p_w)^2 - z]}\ , ( k - 2 p_h)^{\nu}\ , \frac{m_1 ^ 2 - m_2 ^ 2}{m_z^2}\ , + \frac{1}{d_4 ( k^2 - w)}\ , ( k - 2 p_h)^{\nu}\ , \frac{m_h^2 - m_2 ^ 2}{m_w^2 } \nonumber\\ & - & k^{\nu}\ , \left ( \frac{1}{d_1 d_4}\ , + \frac{1}{d_4 d_5 } \right ) + k^{\nu}\ , \frac{1}{d_1 d_4 d_5}\ , ( 4 m_2 ^ 2 - 2 m_w^2 + 2 m_1 ^ 2 + 2 m_h^2 - m_z^2)\ , \nonumber\\ & + & k^{\nu}\ , \frac{(m_2 ^ 2 - m_h^2)(m_1 ^ 2 - m_2 ^ 2)}{m_w^2}\ , \frac{1}{d_1 d_4 d_5}\ , - k^{\nu}\ , \frac{m_2 ^ 2}{m_w^2 d_1 d_5 } \nonumber\\ & + & ( 2 p_h - k)^{\nu}\ , \left [ \frac{3 m_2 ^ 2 + 2 m_w^2 - m_1 ^ 2 - 2 m_h^2}{d_1 d_4 d_5}\ , + \frac{m_1 ^ 2 - m_2 ^ 2}{m_z^2}\ , \frac{1}{d_4 d_5 } \right .
\nonumber\\ & + & \left . \left .
\frac{m_h^2 - m_2 ^ 2}{m_w^2}\ , \left ( \frac{1}{d_1 d_4}\ , + \frac{m_z^2}{d_1
d_4 d_5 } \right ) \right]\ , - ( m_2 \rightarrow m_3 ) \right\}\ , . \label{m10}\end{aligned}\ ] ] it is simple to check now that most terms depending on either of the unphysical squared - masses @xmath91 or @xmath94 cancel among @xmath131 through @xmath132 .
the only two exceptions are the first terms in the curly brackets in the expressions for @xmath133 and for @xmath132 .
the term in @xmath133 contains the integral of @xmath134 $ ] , while the one in @xmath132 contains the integral of @xmath135 $ ] .
however , it is easily found that the absorptive parts of both these integrals are proportional to @xmath103 , and therefore give a vanishing contribution to the amplitude .
this ensures the gauge invariance of the whole computation . in each of eqs .
[ m4 ] to [ m10 ] , therefore , one only has to consider only the terms independent of @xmath91 and of @xmath94 .
those terms are the ones that would have been obtained had the computation been performed directly in the unitary gauge .
some of those terms , however , still yield zero absorptive contributions .
in this section i briefly review the computation of the absorptive parts of the integrals . consider first an integral with @xmath108 in the denominator : @xmath136[(k + p_y)^2 - m_b^2]}\ , , \label{i4}\ ] ] where @xmath137 and @xmath138 are some coefficients .
let @xmath139 .
the integral is divergent . after introducing a feynman parameter @xmath140 , and using dimensional regularization , one obtains @xmath141\ , \left ( \frac{2}{4-d}\ , - \gamma - \ln \frac{\delta}{4 \pi \mu^2 } \right)\ , , \label{result4}\ ] ] where @xmath142 the absorptive part arises when @xmath143 is negative for part of the integration domain of @xmath140 .
this happens when @xmath144 is larger than @xmath145 .
substituting @xmath146 by @xmath147 and integrating over the part of the interval @xmath148 $ ] in which @xmath143 is negative , one gets @xmath149 + s p_h^{\nu } \right\}\ , .
\label{absorptive4}\end{aligned}\ ] ] using this result , one sees that the absorptive parts of some of the integrals in the previous section are proportional to @xmath103 , and therefore irrelevant .
this is true not only of the two gauge - dependent integrals which did not cancel out , as i pointed out in the last paragraph of the previous section ; it is also true of the absorptive parts of the integrals of @xmath150 ( which arises in @xmath132 ) , of @xmath151 ( in @xmath133 ) , of @xmath152 ( in @xmath153 ) , and of @xmath154 ( in @xmath155 ) . eliminating all of these
, it is seen that the only remaining integrals of the type @xmath156 either have denominator @xmath157 , or have denominator @xmath158 .
the first ones have an absorptive part if @xmath159 , and therefore @xmath27 can decay to @xmath160 , and the second ones have an absorptive part if @xmath161 , _ i.e. _ , if @xmath162 is possible
. now consider the integrals with @xmath109 in the denominator . in our problem , they are all of the form @xmath163 [ ( k - p_h)^2 - m_c^2]}\ , , \label{i6}\ ] ] where @xmath137 and @xmath138 are coefficients . specifically , for the integral of @xmath164 present in diagram 7 we have @xmath165 , @xmath166 and @xmath167 ; for the integral of @xmath168 present in diagram 8 we have @xmath169 , @xmath170 and @xmath171 ; for the integral of @xmath172 present in diagram 9 it is @xmath173 , and @xmath171 ; and , for the integral of @xmath174 present in diagram 10 , it is @xmath175 , @xmath176 and @xmath177 .
the integral @xmath178 is finite and , eliminating its part proportional to @xmath103 , one obtains @xmath179 where @xmath180 and @xmath181 in a positive infinitesimal quantity .
using now the fact that the imaginary part of @xmath182 is @xmath183 , and then integrating over the feynman parameter @xmath184 , thereby getting rid of the dirac @xmath185 function , one gets @xmath186 here , the integral must be performed over that part of the interval @xmath148 $ ] for which the two values of @xmath184 defined by the condition @xmath187 are positive and smaller than 1 . that sub - interval of @xmath148 $ ] must be found for each of the four integrals of the type @xmath178 individually .
it is larger or smaller according to whether some cuts can be made in the corresponding feynman diagram : namely , whether @xmath188 or not , and whether @xmath189 or not . in any case , the integral over @xmath140 can be performed analytically , and it yields a rather complicated logarithmic function .
we now have all the ingredients needed to compute the asymmetry .
the asymmetry is equal to @xmath190 times a mixing - matrix factor , studied in section 2 , which should be of order @xmath191 to @xmath192 , times the sum @xmath78 of all absorptive parts , which itself includes a suppression factor @xmath193 .
now , one should note that the absorptive parts of diagrams 1 , 2 , 3 and 6 all diverge when @xmath30 ( or @xmath31 ) approach @xmath29 .
this is simply because in those diagrams one has a scalar @xmath2 ( or @xmath3 ) propagating with momentum @xmath102 such that @xmath99 .
those divergences do not cancel in the absorptive parts , because the specific values of those absorptive parts depend on different parameters : in diagram 1 , on the decay width @xmath194 , in diagram 2 , on the decay width @xmath195 , in diagram 3 , on the decay width @xmath4 , and , in diagram 6 , on the decay width @xmath162 or @xmath196 .
of course , we know that these divergences are not genuine , they might be eliminated by a proper treatment in which one would take into account the finite width of the propagating @xmath2 or @xmath3 , and besides , from a different line of reasoning @xcite , one knows anyway that cp violation disappears and @xmath0 loses its meaning when the masses of any two of the three neutral scalars become equal . a proper treatment of these divergences at @xmath197 or @xmath198 would lead me far astray , and therefore i simply avoided , in general , considering the region of the parameter space in which either @xmath30 or @xmath31 are close to @xmath29 . avoiding those regions in which the present approximation loses its validity , i find that the sum @xmath78 of all absorptive parts is typically of order of magnitude @xmath199 or @xmath200 .
a few examples are presented in the following table . [ cols="^,^,^,^,^",options="header " , ]
it is worth remarking that the total absorptive part @xmath78 is always the final result of substantial cancellations among the absorptive parts , with different signs , of the various individual diagrams .
in this paper i have presented a model calculation of a cp - violating asymmetry in the two - scalar - doublet model .
the asymmetry chosen has been the different decay rates for @xmath4 and for @xmath5 .
i believe this to be a quite interesting place to look for cp violation in the scalar sector , even if the present calculation turns out not to be very relevant .
this might happen mainly because i have taken into account only one source of cp violation , @xmath0 in the gauge interactions of the scalars , while i neglected further sources of cp violation , in the cubic scalar interactions and in the yukawa interactions with the fermions . even within the context of the approximation that i have used , in which the decay proceeds essentially because of a tree - level vertex , while the cp violation arises because of the interference of that tree - level vertex with one - loop diagrams with absorptive parts , those further sources of cp violation will in principle lead to extra contributions to the total asymmetry .
my interest here has been to ilustrate the specific way in which @xmath0 arises in the computation of a cp asymmetry .
i observed that there is a kind of balance between the cp asymmetry and the decay rate in this specific case but only if the only source of cp violation is taken to be @xmath0 , with a large asymmetry being possible only when the decay rate is small , and vice - versa .
i found that the asymmetry can attain values of order @xmath200 .
these values would increase or decrease if the interference with other sources of cp violation in this mode were constructive or destructive . because of the presence of gauge bosons in the internal lines of the one - loop diagrams that i had to compute , i found it convenient to check the gauge invariance of the whole calculation .
i checked that the fictitious masses that appear in the propagators of the @xmath32 and of the @xmath17 , and of the corresponding goldstone bosons and ghosts , in a general t hooft gauge , lead to gauge - dependent absorptive parts for the individual diagrams , which however cancel out when all the diagrams which lead to cp violation proportional to @xmath0 are considered .
this constitutes a good check that one did not omit any diagram .
i thank lincoln wolfenstein for suggesting to me the computation in this work , and for reading the manuscript .
ling - fong li and martin savage made useful technical suggestions .
this work was initiated during visits to the stanford linear accelerator center , and to the institute for nuclear theory at the university of washington , the hospitality of which i acknowledge .
this work was supported by the united states department of energy , under the contract de - fg02 - 91er-40682 .
99 a. d. sakharov , jetp lett .
* 5 * , 24 ( 1967 ) .
m. b. gavela , p. hernndez , j. orloff , and o. pne , mod .
lett . a * 9 * , 795 ( 1994 ) ; + m. b. gavela , p. hernndez , j. orloff , o. pne , and c. quimbay , cern report cern - th-7263 - 94 , 1994 ( unpublished ) .
l. mclerran , phys .
lett . * 62 * , 1075 ( 1989 ) ; + n. turok and j. zadrozny , nucl .
b * 358 * , 471 ( 1991 ) ; + m. dine , p. huet , r. singleton jr . , and l. susskind , phys .
b * 259 * , 351 ( 1991 ) ; + l. mclerran , m. shaposhnikov , n. turok , and m. voloshin , phys . lett .
b * 256 * , 451 ( 1991 ) ; + a. cohen , d. kaplan , and a. nelson , phys .
b * 263 * , 86 ( 1991 ) .
a. mndez and a. pomarol , phys .
b * 272 * , 313 ( 1991 ) .
g. cveti , m. nowakowski , and a. pilaftsis , phys .
b bf 301 , 77 ( 1993 ) .
g. cveti , phys .
d * 48 * , 5280 ( 1993 ) .
b. grzadkowski and j. f. gunion , phys . lett .
b * 294 * , 361 ( 1992 ) ; + a. soni and r. m. xu , phys .
d * 48 * , 5259 ( 1993 ) ; + d. chang and w .- y .
keung , phys .
b * 305 * , 261 ( 1993 ) ; + a. pilaftsis and m. nowakowski , int .
j. mod .
phys . a * 9 * , 1097 ( 1994 ) .
silva and l. lavoura , carnegie - mellon university report no .
cmu - hep94 - 13 ( april 1994 ) , to be published in the physical review d. c. jarlskog , phys .
lett . * 55 * , 1039 ( 1985 ) .
figure 1 : the vertices needed for the computation of the diagrams in figure 2 , and the feynman rules for their values .
all the particles are incoming particles , and all the momenta are incoming momenta . the indices
@xmath105 , @xmath201 and @xmath202 may be @xmath192 , @xmath203 or @xmath204 .
the meaning of the parameters @xmath91 and @xmath94 is explained in the caption of figure 3 .
other vertices which might at first sight be relevant , like @xmath205 and @xmath206 , do not exist .
figure 2 : the one - loop diagrams with external incoming particles @xmath22 , @xmath52 and @xmath1 , which lead to cp violation upon interfering with the tree - level diagram , if the integrals have absorptive parts . in each case , the @xmath32 and the @xmath17 in internal lines may be substituted by the corresponding goldstone bosons @xmath90 and @xmath40 , respectively ; in the first two diagrams , they may be substituted by ghost loops as well .
figure 3 : the propagators relevant for the computation of the diagrams in figure 2 .
@xmath91 and @xmath94 are unphysical parameters with dimension of squared mass , which also arise in the ghost vertices in figure 1 .
the final physical results are independent of both @xmath91 and @xmath94 . | in a general two - scalar - doublet model without fermions , there is a unique source of cp violation , @xmath0 , in the gauge interactions of the scalars .
it arises in the mixing of the three neutral physical scalars @xmath1 , @xmath2 and @xmath3 .
cp violation may be observed via different decay rates for @xmath4 and @xmath5 ( or , alternatively , for @xmath6 and @xmath7 depending on which decays are kinematically allowed ) .
i compute the part of those cp - violating decay - rate differences which is proportional to @xmath0 .
the cp - invariant final - state - interaction phase is provided by the absorptive parts of the one - loop diagrams .
i check the gauge invariance of the whole calculation .
-15 mm |
let us consider a random sequence @xmath27 of total length @xmath28 with alphabet @xmath29 .
each position of @xmath27 is independently assigned the letter @xmath30 , @xmath31 , @xmath32 , or @xmath33 with corresponding probabilities @xmath34 , @xmath35 , @xmath36 and @xmath37 ( @xmath38 ) .
a word of length @xmath13 has the generic form of @xmath39 , where @xmath40 is the letter at the @xmath41-th position .
the total number of such words is @xmath42 .
this number exceeds @xmath28 when @xmath13 increases to order @xmath43 , therefore most of the words of length @xmath44 will never appear in @xmath27
. then what is the probability @xmath45 of a particular word @xmath46 being a maw of sequence @xmath27 ? for @xmath46 to be a maw , it must not appear in @xmath27 but its two subwords of length @xmath47 , @xmath48 and @xmath49 , must appear in @xmath27 at least once , as demonstrated schematically in fig .
[ fig : modelvsrandom](a ) .
we define the core of the maw @xmath46 as the substring @xmath50 which must appear in @xmath27 at least twice , except for the special case of @xmath51 where the @xmath52 and @xmath52 overlap ( see appendix a in _ supplemental information _ ) .
the core must immediately follow the letter @xmath53 at least once and it must also be immediately followed by the letter @xmath54 at least once . similarly ,
if @xmath55 immediately follows the letter @xmath53 , it must not be immediately followed by the letter @xmath54 .
we can construct @xmath56 subsequences of length @xmath13 from @xmath27 , say @xmath57 .
neighboring subsequences are not fully independent as there is an overlap of length @xmath58 between @xmath59 and @xmath60 with @xmath61 .
however , for @xmath62 two randomly chosen subsequences of length @xmath13 from the random sequence @xmath27 have a high probability of being completely uncorrelated .
we can thus safely neglect these short - range correlations , and consequentially the probability of word @xmath46 being a maw is expressed as @xmath63^{n - l+1 } \nonumber \\ & & - \bigl\{\bigl [ 1 - \omega({\bf w}^{(p ) } ) \bigr]^{n - l+2 } + \bigl [ 1 - \omega({\bf w}^{(s ) } ) \bigr]^{n - l+2 } \nonumber \\ & & \quad - \bigl [ 1 - \omega({\bf w}^{(p ) } ) - \omega({\bf w}^{(s ) } ) + \omega({\bf w } ) \bigr]^{n - l+1 } \bigr\ } \ ; , \label{eq : rdprob}\end{aligned}\ ] ] where @xmath64 ( with @xmath65 being the @xmath41-th letter of @xmath46 ) is the probability of a randomly chosen subsequence of length @xmath13 from @xmath27 to be identical to the word @xmath46 , while @xmath66 and @xmath67 are , respectively , the probabilities of a randomly chosen subsequences of length @xmath47 from @xmath27 being identical to @xmath52 and @xmath68 .
summing over all the @xmath42 possible words of length @xmath13 , we obtain the expected number @xmath69 of maws of length @xmath13 for a random sequence @xmath27 of length @xmath28 : @xmath70 where the summation is over all the @xmath71 combinations of the two terminal letters @xmath72 and over all the possibilities with which the letters @xmath30 , @xmath31 , @xmath32 , @xmath33 , may appear in the core a total number of times equal to respectively @xmath73 , @xmath74 , @xmath75 , and @xmath76 .
( a ) illustration of the properties of a minimal absent word and its subwords .
( b ) comparison between the length distribution predicted by eq .
[ eq : mawnum ] and the number of maws calculated for one instance of a random genome of @xmath77 mbp with uniform nucleotide distribution and with @xmath78 gc content .
, title="fig : " ] ( a ) ( a ) illustration of the properties of a minimal absent word and its subwords .
( b ) comparison between the length distribution predicted by eq .
[ eq : mawnum ] and the number of maws calculated for one instance of a random genome of @xmath77 mbp with uniform nucleotide distribution and with @xmath78 gc content .
, title="fig : " ] ( b ) in the simplest case of maximally random sequences , namely @xmath79 , eq .
( [ eq : mawnum ] ) reduces to @xmath80 \ ; .
\label{eq : mawnummaxrand}\end{aligned}\ ] ] we have checked by numerical simulations ( see fig . [
fig : modelvsrandom]b ) that eq .
[ eq : mawnum ] and eq . [ eq : mawnummaxrand ] indeed give excellent predictions of the number maws as a function of their length in random sequences .
we define a predicted minimum and a predicted maximum of the support of the bulk ( @xmath24 and @xmath16 ) as the two values of @xmath13 such that @xmath81 . in the general case , requiring that @xmath82 we obtain @xmath83 , with the shortest length @xmath24 such that @xmath84 is closest to one , while in the other limit we obtain that @xmath85 , with the longest length @xmath16 being @xmath86 the bulk distribution is therefore centered around lengths of order @xmath87 . in the case of maximally random sequences , we can obtain the lower limit analytically and also the first correction to ( [ eq : maxl - gen ] ) , as @xmath88 and @xmath89 the above definition of @xmath16 is good enough for our purposes , and @xmath24 is also a good predictor for @xmath2 ( see below ) .
a more refined predictor for @xmath24 is discussed in appendix b in the _ supplemental information_.
we now describe a protocol for constructing random genome by a iterative copy paste mutation scheme that qualitatively reproduces the tail behavior observed for most of real genomes .
the model is in principle similar to @xcite but differs in the details of the implementation .
the starting point is a string of nucleotides chosen independently at random with a length @xmath90 . at each iteration , we chose two positions @xmath41 and @xmath91 uniformly at random on the genome and a length @xmath4 from a poisson distribution with mean @xmath92 .
we copy the sequence between @xmath41 to @xmath93 and insert it between positions @xmath91 and @xmath94 , thus increasing the genome size by @xmath4 .
we then randomly alter a fraction @xmath95 of nucleotides in the genome , choosing the positions uniformly at random and the new letters from an arbitrary distribution that can be tuned to adjust the gc content .
the process is repeated until the genome reaches the desired length . in this model
, @xmath92 represents the typical size of region involved in a translocation in the genome and @xmath95 corresponds to the expected number of mutations between such events .
we observed that the length @xmath90 or the content of the initial string is unimportant provided that it is much shorter than the final genome size .
the exact value of @xmath92 given a constant @xmath96 ration only affects the tail of the distribution far away from the bulk .
we have also checked by simulations that using different distributions for the choice of @xmath4 does not affect the results qualitatively ( see fig .
s1 in _ supplemental information _ ) .
a low ratio @xmath96 generates genomes with tail maws while higher values cause them to only have bulk maws as in random texts discussed above ( fig .
[ fig : tailmodelresult ] ) .
this is in agreement with the observations for viruses in fig .
[ fig : absentwordsplot ] : _ hepatitis b _ and _ h5n1 _ are viruses that replicate using an error prone reverse transcription and the maw distributions for their genomes lack the tail .
in contrast , _ human herpers 5 _ virus and the _ cafeteria roenbergensis _ virus are dna viruses that use the higher fidelity dna replication mechanism and their genomes clearly have tail maws
. ratio .
the curves were obtained using @xmath97 and a constant @xmath98 .
[ fig : tailmodelresult ] ]
equations ( [ eq : lmin ] ) and ( [ eq : minl - gen ] ) can be used to estimate the length of the shortest absent words .
[ fig : lminallgenomes ] compares the prediction of of the simplest estimate in eq .
[ eq : lmin ] to the length of the shortest maw for a large set of genomes .
the estimator eq .
[ eq : lmin ] is expected to be most accurate only for genomes with neutral gc content .
the figure reveals that genomes of comparable sizes typically vary in their @xmath2 by about @xmath99 nt , and that our estimator captures very well the upper values in this distribution . using eq .
[ eq : minl - gen ] only improves the predictions for genomes with much biased gc content ( @xmath100 or less ) and leads to results in line with the earlier published estimator by wu et al .
@xcite , see appendix c and table s1 in _
supplemental information_. this analysis shows that , contrary to the conclusion of @xcite , there is no need to invoke a biological mechanism to explain the length of the shortest maw ; it is instead a property of rare events when sampling from a random distribution .
indeed , the estimator eq .
[ eq : minl - gen ] gives the length of the shortest human maw as @xmath101 and not @xmath15 , only one nucleotide away from the correct answer ( @xmath5 ) . .
[ fig : lminallgenomes ] ]
the human herpes virus 5 , a double stranded dna virus with a linear genome of @xmath102 kbp , has four very long maws with lengths of @xmath103 nt for two of them , and @xmath104 nt for two others , all other maws being much shorter ( @xmath105 nt or less ) . the cores of these four maws come from three regions , two of them located at the very beginning and very end of the genome , and the third at position @xmath106 kbp made up of the juxtaposition of the reverse complements of the two others .
these regions are annotated as repeated and regulatory .
based on the ncbi blast webservice , these particular sequences are highly conserved ( @xmath107 or more ) in numerous strains of the virus and do not seem to have homologues in any other species : the closely related _ human herpes virus 2 _ shows sequences with no more than @xmath108 similarities to these maw cores . in _ e.
coli _ and _ e. faecalis _ , the @xmath109 longest maws with lengths between @xmath110 and @xmath111 nt all originate from rrna regions : a set of genes present in a few copies made of highly conserved regions with minor variations between the copies . for yeast ,
the four longest maws ( @xmath112 nt ) originate from the two copies of rrna rdn37 on chromosome xii , another four ( @xmath113 nt ) are caused by @xmath114 copies of the region containing pau1 to vth2 on chromosome x and pau4 to vth2 on chromosome ix and two more ( @xmath115 nt ) originate from the copies of gene yrf on chromosome vii and xvi ( yrf is present in at least @xmath116 copies on the yeast genome ) .
we performed an extensive search for all occurrences of the cores of every maw found in the organisms mentioned above and considered the density of maw cores along the genome for maws in the tail ( see fig .
s2 in _ supplemental information _ , ) . except for the human herpes virus 5 that only has few maws which cluster in the repeated segments discussed above
, maw cores do not appear to be preferentially located in any specific regions on the genome scale . a more detailed analysis ( fig .
[ fig : distribmaw2 ] ) however reveals that , while cores of maws from the bulk appear uniformly distributed , those from the tail cluster downstream of ends of annotated coding dna sequences ( cds ) ( i.e. in the @xmath26 utrs and terminator sequences ) . a similar yet weaker effect
can be observed upstream of the start of annotated cds ( the @xmath1 utr ) .
+ by definition , a maw core corresponds to a repeated region on the genome immediately surrounded by nucleotides varying between the copies .
exact repeated regions lead to only a few maws with cores corresponding to that repeat . introducing a few random changes in such regions
creates more but shorter maws , the cores of which are the sub - strings common to two or more regions .
a high density of maw cores in a family of regions such as the utrs thus indicates that they share a limited set of building blocks , implying a similar set of evolutionary constraints or a common origin .
we now consider the lengths of the longest maw found in the genomes of numerous organisms and viruses . we observe that this length generally lies between @xmath117 and a length of about @xmath118 of the genome size ( fig .
[ fig : maxlength ] ) .
viruses are the class showing the largest spread in the length of their longest maws .
many viruses are close to @xmath16 , particularly for those with shorter genomes , confirming our previously mentioned observation that some lack the tail and are thus closer to random texts .
nevertheless , a few viral genomes have maws longer than @xmath118 of the genome length , _
i.e. _ proportionally longer than in any living organism .
the figure suggests that bacteria have on average slightly longer maws than archaea , but overall no clear distinctions between the four types of genomes can be noted based on the length of maws alone , suggesting that the mechanisms behind evolution of all organisms and viruses influence the maws distribution in the same way . a more detailed analysis of the data presented in fig .
[ fig : distribmaw2 ] shows that organisms have the longest maws closest to the lower bound of the tail ( @xmath16 ) or the observed upper bound of @xmath118 of the genome length share some common traits . in bacteria ,
the @xmath119 genomes having their longest maws closest to @xmath16 are two strains from the _ buchnera aphidicola _ species , two strains of _ candidatus carsonella ruddii _
, one _ candidatus phytoplasma solani _ and one _ bacteroides uniformis_. while the last is a putative bacterial species living in human feces @xcite , the five other species are intra - cellular symbiotic or parasitic gammaproteobacteria in plants or insects @xcite . among eukaryotes ,
the same analysis gives us _ plasmodium gaboni _
, an agent responsible for malaria @xcite , a species of _ cryptosporidium _ , another family of intracellular parasites found in drinking water and _ chromera velia _ , a photosynthetic organism from the same apicomplexa phylum as plasmodium , which is remarkable in this class for its ability to survive outside a host and is of particular interest for studying the origin of photosynthesis in eukaryotes @xcite . for archaea
, we find _ candidatus parvarchaeum acidiphilum _ and _ c. p. acidophilus _ , which are two organisms with short genomes ( @xmath120 and @xmath121 kbp ) living in low ph drainage water from the richmond mine in nothern california @xcite , and an uncultivated hyperthermophilic archaea `` scgc aaa471-e16 '' of the aigarchaeota phylum @xcite .
additionally , we searched for maws in @xmath122 human mithocondrial genomes with lengths between @xmath123 and @xmath124 bp and found that the longest maws are only @xmath125 or @xmath6 nt long , while @xmath126 for these genomes . among bacteria having their longest maw close to 10% of the genome length , we find several strains of _ e. coli _ , _ francisella tularensis _ , _ shewanella baltica _ , _ methylobacillus flagellatus _ , _ xanthomonas oryzae _ and a species of the _ wolbachia _ genus .
all of these are facultative or obligatory aerobes and are a lot more widespread than the bacteria listed in the previous paragraph .
at least the four first species are free - living and commonly cultured in labs .
_ x. oryzae _ is a pathogen affecting rice residing in the intercellular spaces .
_ wolbachia _ species are a very common intracellular parasites or symbiotes living in arthropods . among eukaryotes , in addition to human and mouse , we find _ dictyostelium discoideum _ , an organism living in soil that changes from uni- to multi - cellular during its life cycle , and _ thalassiosira pseudonana _ a unicellular algea commonly found in marine phytoplankton .
finally , archaea with the longest maws are _ methanococcus voltae _
, a mesophilic methanogen , _ halobacterium salinarum _ , a halophilic marine obligate aerobic archeon also found in food such as salted pork or sausages , and _ halalkalicoccus jeotgali _ , another halophilic archeon isolated from salted sea - food@xcite . to summarize
, it appears from these results that development in specific environmental niches that offer rather stable conditions , and particularly the inside of the cell of another organism , is , albeit with some exceptions , associated with short maws . on the other hand ,
aerobic life and widespread presence in changing and diverse environments such as soil , sea water or food is generally associated with long maws .
intracellular organisms have a well known tendency to reduce the size of their genomes and increase error rate for dna replication due to elimination of error correction mechanisms @xcite .
this translates into a high value for @xmath95 in our random model for the tail and explains why their longest maws are short .
in particular , we note that _ b. aphidicola _ ( genome size of @xmath127 kbp ) , brought out by our analysis , is known to have the highest mutation rate among all prokaryotes and indeed its longest maw is as short as @xmath128 nt @xcite . as for organisms specialized for a niche environment
, one may hypothesize that proliferation speed is more important than replication fidelity @xcite .
reasons for a widespread presence in changing environments to increase the length of the longest maws are less clear .
multicellular eukaryotic organisms do not show particularly high dna replication fidelity @xcite .
the fidelity of _ e. coli _ is fairly good at about 3.5x the one of human germline @xcite , but it is unclear if this is enough to make it stand out from other bacteria . we speculate that soil , sea water or food , which are likely to contain many types of microorganisms , may favor species more likely to undergo horizontal gene transfers , thus increasing the rate of translocation events and decreasing @xmath95 in our model , while leaving the _ per generation _ mutation rate unchanged . .
the dashed line represents the estimator in eq .
[ eq : lmax ] and the dotted line @xmath129 .
[ fig : maxlength ] ]
we have proposed a two - parts model explaining the unusual shape of the length distribution of minimal absent words ( maws ) in genomes .
the first part of the model quantitatively reproduces the bulk of the distribution by considering the genome as a random text with random and independent letters .
the second part is a stochastic algorithm grounded in basic principles of how genomes evolve , through translocation events and mutations , that qualitatively reproduces the behavior in the tail .
our theory provides an estimator for the length of the shortest maws that is remarkably simple and captures well the global trend observed in large numbers of genomes from all sorts of organisms and viruses .
considerations about the longest maw in a genome reveal sets of organisms sharing common high - level features such as the type of environment they live in .
we have shown arguments for believing that organisms and viruses having few tail maws do so because of a low replication fidelity .
why some organisms such as _ e. coli _ have long tail maws is less clear and replication fidelity alone does not seem to be a sufficient reason . finally , we have introduced the concept of maw cores , sequences present on the genome that tell us about what causes the existence of their parent maw .
we have shown that , while cores from bulk maws do not seem biologically relevant , cores from tail maws cluster in regions of the genome with special roles , namely ribosomal rnas and untranslated regions surrounding coding regions of genes , a feature that can not be explained by a stochastic protocol that ignores the biological roles of the strings it manipulates .
viral and bacterial genomes were downloaded under the form of the `` all.fna '' archives from the `` genomes / viruses/ '' and `` genomes / bacteria '' from the ncbi database on 17 - 18 may 2015 respectively .
the norway spruce s genome was downloaded from the `` spruce genome project''@xcite homepage and the yeast genome strain 288c@xcite from saccharomyces genome database .
genomes of other eukaryotes and archaeas were downloaded from the ncbi database at several different dates over the period may - june 2015 .
the human mithocondrial genomes were downloaded from the human mitochondrial genome database ( mtdb)@xcite in early september 2015 .
all maws were computed using the software provided by pinho et al . in @xcite .
the software was run taking into account the reverse - complementary strand ( -rc command line switch ) and requesting maws ( -n command line option ) with length up to five million nucleotides , i.e. much longer than the expected length of the longest maws .
the search for maws was performed on commodity desktop computers for all but the human , mouse , and norway spruce genomes , for which the computer `` ellen '' from the center for high performance computing ( pdc ) at kth was used .
localization of occurrences of maw cores on the genomes for figures [ fig : distribmaw2 ] and s2 was done by aligning these subwords to their respective genomes using bowtie2 @xcite , allowing only strict alignments ( command line option -v 0 ) .
identical cores from different maws are counted independently in the coverage .
this research is supported by the swedish science council through grant 621 - 2012 - 2982 ( ea ) , by the academy of finland through its center of excellence coin ( ea ) , and by the natural science foundation of china through grant 11225526 ( hjz ) .
ea thanks the hospitality of kitpc and hjz thanks the hospitality of kth .
the authors thank profs rdiger urbanke and nicolas macris , and dr .
franoise wessner for valuable discussions , and pdc , the center for high performance computing at kth , for access to computational resources needed to analyze large genomes .
amir , a , zeisel , a , zuk , o , elgart , m , stern , s , shamir , o , turnbaugh , j , soen , y , & shental , n. ( 2013 ) high - resolution microbial community reconstruction by integrating short reads from multiple 16s rrna regions . .
chatterjee , s , koslicki , d , dong , s , innocenti , n , cheng , l , lan , y , vehkaper , m , skoglund , m , rasmussen , l. k , aurell , e , & corander , j. ( 2014 ) sek : sparsity exploiting k - mer - based estimation of bacterial community composition .
, 24232431 .
fofanov , y , fofanov , y , & pettitt , b. ( 2002 ) _ counting array algorithms for the problem of finding appearances of all possible patterns of size @xmath130 in a sequence_. ( w.m . keck center for computational and structural biology ) .
innocenti , n , golumbeanu , m , dhroul , a. f , lacoux , c , bonnin , r. a , kennedy , s. p , wessner , f , serror , p , bouloc , p , repoila , f , & aurell , e. ( 2015 ) whole genome mapping of 5 ends in bacteria by tagged sequencing : a comprehensive view in _ enterococcus faecalis_. , 10181030 .
quaglino , f , zhao , y , casati , p , bulgari , d , bianco , p. a , wei , w , & davis , r. e. ( 2013 ) _candidatus phytoplasma solani_ , a novel taxon associated with stolbur and bois noir related diseases of plants .
ijs0 .
van ham , r. c. h. j , kamerbeek , j , palacios , c , rausell , c , abascal , f , bastolla , u , fernndez , j. m , jimnez , l , postigo , m , silva , f. j , tamames , j , viguera , e , latorre , a , valencia , a , morn , f , & moya , a. ( 2003 ) reductive genome evolution in _
buchnera aphidicola_. , 581586 .
moore , r. b , obornk , m , janoukovec , j , chrudimsky , t , vancov , m , green , d. h , wright , s. w , davies , n. w , bolch , c. j , heimann , k , et al .
( 2008 ) a photosynthetic alveolate closely related to apicomplexan parasites . , 959963 .
baker , b. j , comolli , l. r , dick , g. j , hauser , l. j , hyatt , d , dill , b. d , land , m. l , verberkmoes , n. c , hettich , r. l , & banfield , j. f. ( 2010 ) enigmatic , ultrasmall , uncultivated archaea .
, 88068811 .
rinke , c , schwientek , p , sczyrba , a , ivanova , n. n , anderson , i. j , cheng , j .- f , darling , a , malfatti , s , swan , b. k , gies , e. a , dodsworth , j. a , hedlund , b. p , tsiamis , g , sievert , s. m , liu , w .-
t , eisen , j. a , hallam , s. j , kyrpides , n. c , stepanauskas , r , rubin , e. m , hugenholtz , p , & woyke , t. ( 2013 ) insights into the phylogeny and coding potential of microbial dark matter .
, 431437 .
roh , s. w , nam , y .- d , chang , h .-
w , sung , y , kim , k .-
h , oh , h .-
m , & bae , j .- w . ( 2007 ) _ halalkalicoccus jeotgali _ sp . nov . , a halophilic archaeon from shrimp jeotgal , a traditional korean fermented seafood .
, 22962298 .
nystedt , b , street , n. r , wetterbom , a , zuccolo , a , lin , y .- c , scofield , d. g , vezzi , f , delhomme , n , giacomello , s , alexeyenko , a , vicedomini , r , sahlin , k , sherwood , e , elfstrand , m , gramzow , l , holmberg , k , hallman , j , keech , o , klasson , l , koriabine , m , kucukoglu , m , kaller , m , luthman , j , lysholm , f , niittyla , t , olson , a , rilakovic , n , ritland , c , rossello , j. a , sena , j , svensson , t , talavera - lopez , c , theiszen , g , tuominen , h , vanneste , k , wu , z .-
q , zhang , b , zerbe , p , arvestad , l , bhalerao , r , bohlmann , j , bousquet , j , garcia gil , r , hvidsten , t. r , de jong , p , mackay , j , morgante , m , ritland , k , sundberg , b , lee thompson , s , van de peer , y , andersson , b , nilsson , o , ingvarsson , p. k , lundeberg , j , & jansson , s. ( 2013 ) the norway spruce genome sequence and conifer genome evolution .
, 579584 .
engel , s. r , dietrich , f. s , fisk , d. g , binkley , g , balakrishnan , r , costanzo , m. c , dwight , s. s , hitz , b. c , karra , k , nash , r. s , weng , s , wong , e. d , lloyd , p , skrzypek , m. s , miyasato , s. r , simison , m , & cherry , j. m. ( 2014 ) the reference genome sequence of _ saccharomyces cerevisiae _ : then and now .
, 389398 . | minimal absent words ( maw ) of a genomic sequence are subsequences that are absent themselves but the subwords of which are all present in the sequence . the characteristic distribution of genomic maws as a function of their length has been observed to be qualitatively similar for all living organisms , the bulk being rather short , and only relatively few being long .
it has been an open issue whether the reason behind this phenomenon is statistical or reflects a biological mechanism , and what biological information is contained in absent words . in this work
we demonstrate that the bulk can be described by a probabilistic model of sampling words from random sequences , while the tail of long maws is of biological origin .
we introduce the novel concept of a core of a minimal absent word , which are sequences present in the genome and closest to a given maw .
we show that in bacteria and yeast the cores of the longest maws , which exist in two or more copies , are located in highly conserved regions the most prominent example being ribosomal rnas ( rrnas ) .
we also show that while the distribution of the cores of long maws is roughly uniform over these genomes on a coarse - grained level , on a more detailed level it is strongly enhanced in 3 untranslated regions ( utrs ) and , to a lesser extent , also in 5 utrs .
this indicates that maws and associated maw cores correspond to fine - tuned evolutionary relationships , and suggest that they can be more widely used as markers for genomic complexity .
enomic sequences are texts in languages shaped by evolution .
the simplest statistical properties of these languages are short - range dependencies , ranging from single - nucleotide frequencies ( gc content ) to @xmath0-step markov models , both of which are central to gene prediction and many other bioinformatic tasks @xcite .
more complex characteristics , such as abundances of @xmath0-mers , sub - sequences of length @xmath0 , have applications to classification of genomic sequences @xcite , and _ e.g. _ to fast computations of species abundancies in metagenomic data @xcite
. the reverse image of words present are absent words ( aws ) , subsequences which actually can not be found in a text . in genomics
the concept was first introduced around 15 years ago for fragment assembly @xcite and for species identification @xcite , and later developed for inter- and intra - species comparisons @xcite as well as for phylogeny construction @xcite .
a practical application is to the design of molecular bar codes such as in the tagrna - seq protocol recently introduced by us to distinguish primary and processed transcripts in bacteria @xcite .
short sequences or tags are ligated to transcript @xmath1 ends , and reads from processed and primary transcripts can be distinguished _ in silico _
after sequencing based on the tags . for this to be possible
it is crucial that the tags do not match any subsequence of the genome under study , _
i.e. _ that they correspond to absent words . in a further recent study
we also showed that the same method allows to separate true antisense transcripts from sequencing artifacts giving a high - fidelity high - throughput antisense transcript discovery protocol @xcite . in these as in other biotechnological applications there is an interest in finding short absent words , preferably additionally with some tolerance .
minimal absent words ( maw ) are absent words which can not be found by concatenating a substring to another absent word .
all the subsequences of a maw are present in the text .
maws in genomic sequences have been addressed repeatedly @xcite as these obviously form a basis for the derived set of all absent words .
furthermore , while the number of absent words grows exponentially with their length @xcite , because new aws can be built by adding letters to other aws , the number of maws for genomes shows a drastically different behavior , as illustrated below in fig .
[ fig : absentwordsplot](a ) and previously reported in the literature @xcite .
the behavior can be summarized as there being one or more shortest minimal absent word of a length which we will denote @xmath2 , a maximum of the distribution at a length we will denote @xmath3 , and a very slow decay of the distribution for large @xmath4 . in human @xmath2 is equal to @xmath5 , as first found in @xcite , @xmath3 is equal to @xmath6 , there being about @xmath7 billion maws of that length , while the support of the distribution extends to around @xmath8 ( fig .
[ fig : absentwordsplot](c ) and ( d ) ) .
the total number of human maws is about eight billion . as already found in @xcite the very end of the distribution
depends on the genome assembly ; for human genome assembly grch38.p2 the three longest maws are @xmath9 , @xmath10 and @xmath11 nt in length .
( a ) ( b ) ( c ) ( d ) several aspects of this distribution are interesting .
first , in a four - letter alphabet there are @xmath12 possible subsequences of length @xmath0 , but in a text of length @xmath13 only @xmath14 subsequences of length @xmath0 actually appear . if the human genome were a completely random string of letters one would therefore expect the shortest maw to be of length @xmath15
the fact that @xmath2 is considerably shorter ( @xmath5 ) is therefore already an indication of a systematic bias , in @xcite attributed to the hypermutability of cpg sites .
we will return to this point below .
more intriguing is the observation that the overwhelming majority of the maws lie in a smooth distribution around @xmath3 , and then a small minority are found at longer lengths .
we will call the first part of the distribution the _ bulk _ and the second the _ tail_. we separate the tail from the bulk by a cut - off @xmath16 which we describe below ; the human @xmath16 is @xmath17 , a typical number for larger genomes , while the _ escherichia coli _ @xmath16 is @xmath18 . using this separation there are about @xmath19 million human tail maws , about @xmath20 of the total , while there are @xmath21 _ e. coli _ tail maws , about @xmath22 of the total .
the effect is qualitatively the same for , as far as we are aware of , all eukaryotic , archeal and baterial genomes analyzed in the literature @xcite , as well as all tested by us .
only a few viruses with short genomes are exceptions to this rule and show only the bulk , see fig .
[ fig : absentwordsplot](c ) .
the questions we want to answer in this work are why the distributions of maws are described by the bulk and the tail .
can these be understood quantitatively ?
do they carry biological information or are they some kind of sampling effects ?
can one make further observations ?
we will show that both the bulk and the tail can be described probabilistically , but in two very different ways .
the bulk of the maw distribution arises from sampling words from finite random sequences and are contained in an interval @xmath23 $ ] , where @xmath16 was introduced above and @xmath24 is a good predictor of @xmath25 , the actual length of the shortest aws . to the best of our knowledge this has not been shown previously , and although our analysis uses only elementary considerations , they have to be combined in a somewhat intricate manner . the distribution of bulk maws , which comprise the vast majority of maws in all genomic sequences , can hence be seen as nothing more than a complicated transformation of simple statistical properties of the sequence .
in fact , excellent results are obtained taking only the single nucleotide composition into account ( fig [ fig : absentwordsplot](b ) ) .
nevertheless , the tail maws are different , and can be described by a statistical model of genome growth by a copy paste mutate mechanism similar to the one presented in @xcite .
we show that the distributions of the tail maws vary , both in the data and in the model .
the human and the mouse maw tail distributions follow approximately a power - law , but this seems to be more the exception than the rule ; bacteria and yeast as well as _
e.g. _ _ picea abies _
( norway spruce ) show a cross - over behavior to a largest maw length . for bacteria
this largest length ranges from hundreds to thousands ; for _ p. abies _ it is around 30 000 ; while for human and mouse the tail maw distribution reaches up to one million , without cross - over behavior ( fig [ fig : absentwordsplot](d ) ) . from the definition ,
any subword obtained by removing letters from the start or end of a maw is present in the sequence . in particular , removing the first and last letters of a maw leads to a subword that is present at least twice , which we denote here as a _ maw core_. maws made of a repeat of the same letter are an exception to this rule as they can have the two copies of their cores overlapping each other , see appendix a in _
supplemental information_. maw cores can be considered as the causes that create the maws and their location on the genome combined with functional information from the annotation tells us about their biological significance . finding all the occurrences on a genome of a given word ( such as a maw core ) is the very common bioinformatic task of alignment , which can be done quickly and efficiently using one of the many software packages available . in bacteria and yeast ,
the cores from the longest maws are predominantly found in regions coding for ribosomal rnas ( rrnas ) , regions present in multiple copies on the genome and under high evolutionary pressure as their sequence determines their enzymatic properties , required for protein synthesis and vital to every living cell . at the global scale
, it appears that maw cores obtained from maws in the bulk are distributed roughly uniformly over the genome while those from the tail cluster in @xmath26 utrs and , to a lesser extent , also in @xmath1 utrs .
these regions are important for post - transcriptional regulation , and thus likely to be under evolutionary pressure similarly to rrnas .
we end this introduction by noting that from a linguistic perspective a language can be described by its list of _ forbidden _ sub - sequences , or the list of its forbidden words @xcite .
minimal forbidden words relate to forbidden words as maws to absent words , and in a text of infinite length the lists of maws and minimal forbidden words would agree .
if there is a finite list of minimal forbidden words the resulting language lies on the lowest level of regular languages in the chomsky hierarchy@xcite , and is hence relatively simple , while a complex set of instructions , such as a genome , is expected to correspond to a more complex language , with many layers of meaning .
such aspects have been exploited in cellular automata theory @xcite and in dynamical systems theory @xcite , and are perhaps relevant to genomics as well .
the present investigation is however focused on properties of texts of finite length , for which minimal forbidden words and maws are quite different . |
stellar bars play an important role in the dynamical evolution of gas in galaxies . by introducing a non - axisymmetric torque
, they produce interesting morphological substructures in the gaseous medium , including a pair of dust lanes at the leading side of the bar , a nuclear ring near the center , and nuclear spirals inside the ring ( e.g. , @xcite ) .
they also transport gas inward which can trigger starbursts in the rings ( e.g. , @xcite ) and if the mass inflow extends all the way to the center , they may help power active galactic nuclei ( agn ) ( e.g. , @xcite ) .
since bar substructures represent a nonlinear response of the gas to a non - axisymmetric gravitational potential , their formation and evolution is best studied using direct numerical simulations .
there have been a number of numerical studies on the gas dynamics in barred galaxies . based on the numerical scheme employed
, they can be categorized largely into two groups : ( 1 ) those using a smoothed particle hydrodynamics ( sph ) technique ( e.g. , @xcite ) and ( 2 ) those using a grid - based algorithm ( e.g. , @xcite ) .
the numerical results from these two approaches do not always agree with each other , at least quantitatively , even if the model parameters are almost identical .
for instance , @xcite using the cmhog code on a cylindrical grid reported that the gas near the corotation regions exhibits complex density features resulting from rayleigh - taylor and/or kelvin - helmholtz instabilities , while these structures are absent in the sph simulations .
in addition , overall shapes and structures of dust lanes and nuclear rings from cmhog simulations are different from sph results . some differences in the numerical results may be attributable to relatively large numerical diffusion of a standard sph method and its inability to handle sharp discontinuities accurately ( e.g. , @xcite ) .
however , after adopting and thoroughly testing the cmhog code as part of this work , we have found it contained a serious bug in the way the gravitational forces due to the bar are added to the hydrodynamical equations .
thus , some of the discrepancies in the flows computed by cmhog and other codes are likely due to this bug .
we discuss this bug and its affect on the results reported in @xcite in section 2.2 . in this paper , we revisit the gas dynamics in barred galaxies using a corrected version of the cmhog code .
our objectives are three - fold .
first , we wish to remedy the errors in @xcite , and to compute the formation of bar substructures with an accurate shock - capturing grid code with the correct bar potential .
second , the morphology , shape , and strength of the bar substructures are likely to depend on the gas sound speed and the shape of the underlying gravitational potential ( e.g. , @xcite ) .
thus , we report new models in which we include a central black hole ( bh ) that greatly affects the gravitational potential in the central regions , and we vary both the bh mass as well as the sound speed to explore the dynamics in various galactic conditions .
third , we exploit advances in computational resources to compute models that have more than an order of magnitude higher resolution than the models in @xcite , with a grid resolution of @xmath4 pc in the central regions .
this allows us to resolve details in the flow in the nuclear regions , in particular the formation of nuclear rings and nuclear spirals .
according to the most widely accepted theory , a nuclear ring forms near the inner lindblad resonance ( ilr ) when there is only one ilr , as the gas outside ( inside ) ilr loses ( gains ) angular momentum and accumulates there , while it forms in between the inner ilr and outer ilr when there are two ilrs ( e.g. , @xcite ) . on the other hand , @xcite argued that the ring formation is more deeply related to the existence of @xmath2-orbits rather than the ilrs .
but , the arguments relying on either ilr or @xmath2-orbits do not take into account the effect of thermal pressure .
therefore , it is important to explore to what extent the concepts of ilr or @xmath2-orbits are valid in describing nuclear rings , especially when the sound speed is large . the formation , shape , and nature of nuclear spirals that may channel the gas to the galaxy centers are also not well understood .
observations using the _ hubble space telescope _ indicate that galaxies having nuclear dust spirals are quite common ( e.g. , @xcite ) . while most of such spirals are trailing , a few galaxies including ngc 1241 and ngc 6902 reportedly possess leading nuclear spirals @xcite .
although the linear theory suggests that leading spirals are expected when there are two ilrs ( e.g. , @xcite ) , they are absent in the numerical models of @xcite computed with the cmhog code , while the sph models of @xcite with self - gravity do form leading spirals .
the sph models suffer from poor spatial resolution at the nuclear regions as most particles gather around the rings . by running high - resolution simulations with a corrected version of cmhog
, we can clarify the issues of the nuclear spiral formation and related mass inflow rates to the galaxy center .
we in this work treat gaseous disks as being two - dimensional , isothermal , non - self - gravitating , and unmagnetized , which introduces a few caveats that need be noted from the outset . by considering an infinitesimally - thin disk , we ignore gas motions and associated dynamics along the direction perpendicular to the disk . by imposing a point symmetry relative to the galaxy center
, our models do not allow for the existence of odd-@xmath5 modes , although this appears reasonable since @xmath6 modes dominate in the problems involving a galactic bar .
in addition , we are unable to capture the potential consequences of gaseous self - gravity and magnetic stress that may not only cause fragmentation of high - density nuclear rings but also affect mass inflow rates to the galaxy center . nevertheless , these idealized models are useful to isolate the effects of the gas sound speed and the mass of a central bh on the formation of bar substructures and mass inflows .
also , these models allow us to correct the results of previous cmhog calculations with incorrect bar forces .
this paper is organized as follows . in section 2 , we describe the galaxy model , model parameters , and our numerical methods . in section 3 , we present the results of simulations for off - axis shocks and nuclear rings .
the detailed properties of nuclear spirals are presented in section 4 . in section 5
, we study the mass inflow rates through the inner boundary obtained from our simulations . in section 6 ,
we conclude with a summary and discussion of our results and their astronomical implications .
lcc cs05bh0 & 5 & 0 + cs05bh0 t & 5 & 0 + cs10bh0 & 10 & 0 + cs15bh0 & 15 & 0 + cs20bh0 & 20 & 0 + cs20bh0 t & 20 & 0 + cs05bh7 & 5 & 4@xmath7 + cs10bh7 & 10 & 4@xmath7 + cs15bh7 & 15 & 4@xmath7 + cs20bh7 & 20 & 4@xmath7 + cs20bh7 t & 20 & 4@xmath7 + cs05bh8 & 5 & 4@xmath8 + cs10bh8 & 10 & 4@xmath8 + cs15bh8 & 15 & 4@xmath8 + cs20bh8 & 20 & 4@xmath8
we consider a uniform , isothermal , infinitesimally - thin , and non - self - gravitating gas disk orbiting in a gravitational potential @xmath9 arising from various components of a barred galaxy . the bar is assumed to rotate about the galaxy center with a fixed pattern speed @xmath10 .
therefore , it is advantageous to solve the dynamical equations in cylindrical polar coordinates ( @xmath11 , @xmath12 ) corotating with the bar in the @xmath13 plane .
the equations of ideal hydrodynamics in this rotating frame are @xmath14 @xmath15 where @xmath16 , @xmath17 , and @xmath0 denote the surface density , velocity in the rotating frame , and the sound speed in the gas , respectively .
the third and fourth terms in the right hand side of equation ( [ eq : mom ] ) represent the coriolis and centrifugal forces , respectively , arising from the coordinate transformation from the inertial to rotating frames .
the velocity @xmath18 in the inertial frame is obtained from @xmath19 . in order to focus on the bar - driven gas dynamics
, we do not consider star formation and the associated gas recycling in the present work .
the external gravitational potential @xmath9 consists of four components : an axisymmetric stellar disk , a spherical bulge , a non - axisymmetric bar , and a central supermassive bh .
appendix [ sec : appen ] describes the specific potential model we employ for each component of the galaxy .
the bar pattern speed is taken to be @xmath20 . without a central bh ,
our galaxy model is similar to those in @xcite and @xcite .
the presence of a bh allows us to explore the effect of central mass concentration on the formation of nuclear spirals ( e.g. , @xcite ) .
the real interstellar gas is multiphase and turbulent , with temperatures differing by a few orders of magnitude ( e.g. , @xcite ) .
for simplicity , we model this highly inhomogeneous gas using an isothermal equation of state with an effective sound speed @xmath0 that includes a contribution due to turbulent motions .
we have calculated 15 different models in which we vary both @xmath0 and the initial mass @xmath21 of the central bh as parameters .
table [ tbl : model ] lists the properties of each calculation . the sound speed is chosen to vary between 5 and @xmath22 .
models with a postfix bh0 do not initially possess a central bh , and are similar to those in @xcite .
models with the postfix bh7 or bh8 have a bh with mass @xmath23 and @xmath24 respectively ; they are analogous to models in @xcite and @xcite . for most models
, we fix the bh mass to its initial value , but we also consider three additional models ( cs05bh0 t , cs20bh0 t , and cs20bh7 t ) in which the bh mass is varied with time according to @xmath25 , where @xmath26 is the mass inflow rate across the inner boundary ( see below ) .
these time - varying @xmath1 models allow us to study the effect of bh growth due to the gas accretion on bar substructures .
figure [ fig : rotcurve ] plots the net circular rotation curves together with a contribution from each component when @xmath27 .
the solid and dashed lines are along the bar major and minor axes , respectively .
the circular velocity is almost flat at @xmath28 in the outer parts . without the bh ,
the rotation curve @xmath29 would rise linearly with @xmath11 close to the center , but the presence of the bh results in @xmath30 for @xmath31 .
this rapid increase of @xmath29 will render the gaseous orbits in the very central regions highly resistant to pressure perturbations , resulting in smaller mass inflow rates than the cases without it , as we show below .
figure [ fig : angfreq ] shows the characteristic angular frequencies , @xmath32 , @xmath33 , and @xmath34 along the bar major and minor axes as solid and dashed lines , respectively . as an angular frequency and @xmath35 as a radial frequency do not apply strictly since closed orbits are in general non - circular . in our models
, however , the bar potential is nearly axisymmetric at @xmath36 , so that @xmath33 and @xmath35 measure the actual frequencies reasonably well in the central parts . ] here , @xmath37 and @xmath38 denote the angular and epicyclic frequencies , respectively .
the horizontal dotted line in each panel represents the bar patten speed of @xmath20 , with the corotation resonance ( cr ) located at @xmath39 for all models . for bh0 models with no bh , the @xmath32 curve peaks at @xmath40 and is equal to @xmath41 at the two ilr with radii of @xmath42 and @xmath43 . because the rotation curve rises steeply toward the center , bh7 and bh8 models have only a single ilr at @xmath44 .
the @xmath32 curve in bh7 models attains its local maximum and minimum at @xmath40 and @xmath45 , respectively . in bh8 models ,
on the other hand , it increases monotonically with decreasing @xmath11 since the bh dominates the gravitational potential .
we will show in section [ sec : nsp ] that the shape of nuclear spirals depends critically on the sign of @xmath46 ( e.g. , @xcite ) . to solve equations ( [ eq : con ] ) and ( [ eq : mom ] ) , we use the two - dimensional grid - based code cmhog in cylindrical geometry @xcite . cmhog implements the piecewise parabolic method in its lagrangian remap formulation @xcite , which is third - order accurate in space and
has very little numerical diffusion ( viscosity ) .
all the runs are carried out in a frame corotating with a bar whose major axis is aligned along the @xmath47-axis ( i.e. , @xmath48 ) , so that the bar potential remains stationary in the simulation domain . by assuming a reflection symmetry with respect to the galaxy center , the simulations were performed on a half - plane with @xmath49 constructed by making a cut along the bar major axis . as mentioned in section [ sec : intro ] , the original version of cmhog used by @xcite contained a serious bug in the way the gravitational forces were added to the hydrodynamic equations .
the cmhog code places a bar potential @xmath50 with the major axis aligned along the @xmath51-axis , calculate the bar forces @xmath52 and @xmath53 , and then transforms them into cylindrical coordinates . in the original version of the code , the incorrect transform relations @xmath54 and @xmath55 were used .
in fact , the correct transformation rule for the azimuthal force should be @xmath56 . with the sign of the azimuthal force reversed ( but the radial force correct ) , the flows in models computed using the original cmhog code behave as if the bar potential were aligned parallel to the @xmath47-axis , but with forces quite different from the intended ones .
other than these force transformations , the complex hydrodynamic algorithms in cmhog are all implemented correctly , and were well tested in the original paper by @xcite .
therefore , previous numerical studies based on cmhog should remain valid as long as they did not adopt the incorrect transformations of the bar forces inherited from @xcite .
unfortunately , the results of @xcite were compromised by a trivial sign error in the coordinate transform relations for the bar forces .
figure 3 compares the results for a typical simulation run with the original and corrected version of cmhog . for this test
, the grid resolution is taken identical to that in @xcite with 251 and 154 zones in the radial and azimuthal directions , respectively .
the figure plots the logarithm of the surface density at @xmath57 myr .
several differences are apparent .
for example , the gas around the corotation resonance ( cr ) at @xmath58 in the left panel is largely evacuated and has corrugated streams linked to the ends of the bar major axis , whereas the cr region is relatively featureless in the right panel .
the nuclear ring in the left panel is fairly smooth , while it is quite clumpy in the right panel .
most importantly , the very central region inside the ring is almost unperturbed in the left panel , while the right panel shows spiral structures in the central region .
these differences suggest that the original cmhog code is unable to properly model the flow in the central regions , especially weak nuclear spirals .
we have also run the same model using other grid - based codes adopting cartesian coordinates , such as tvd @xcite and antares @xcite as well as the particle - based gadget code @xcite , in all of which the bar forces are calculated by taking finite differences of @xmath50 rather than using @xmath59 and @xmath60 directly .
the new version of the cmhog code used in this work gives results which are much more similar to the results of these other codes , which gives us further confidence that the gravitational forces due to the bar are now being treated correctly . to resolve the central regions
accurately , we set up a non - uniform , logarithmically - spaced cylindrical grid with 1024 radial zones extending from @xmath61 to @xmath62 and 480 azimuthal zones covering the half - plane .
this makes the zones approximately square - shaped throughout the grid ( i.e. , @xmath63 ) .
the resulting grid spacing is @xmath64 , @xmath65 , and @xmath66 at the inner radial boundary , @xmath67 where nuclear rings typically form , and the outer radial boundary , respectively .
this increases the resolution in the inner regions by over an order of magnitude , in comparison to the models presented in @xcite .
this level of grid resolution is crucial to resolve nuclear spiral structures within @xmath67 .
we use outflow and continuous boundary conditions at the inner and outer radial boundaries , respectively , while the azimuthal boundaries are periodic .
the gas crossing the inner boundary is considered lost from the simulation domain .
we keep track of the total mass crossing the inner boundary in order to study the mass inflow rates into the galactic nucleus .
each model starts from a uniform disk with surface density @xmath68 that is rotating in force balance with an axisymmetric gravitational potential without a bar . in order to avoid strong transients in the fluid flow caused by a sudden introduction of the bar ,
we slowly introduce the bar potential over one bar revolution time of @xmath69 myr .
this is accomplished by increasing the bar central density @xmath70 linearly with time and decreasing the bulge central density @xmath71 , while keeping the net central density @xmath72 fixed .
this ensures that the shape of the total gravitational potential @xmath9 , when averaged along the azimuthal direction , is unchanged with time .
all the models are run until 500 myr .
this corresponds to @xmath73 and 10 orbits at the inner and outer radial boundaries , respectively , for bh8 models with @xmath27 .
we take model cs05bh0 with @xmath74 and no bh as our standard model .
the overall evolution of other models with different @xmath0 and @xmath1 are qualitatively similar , although the properties of the nuclear features that form differ considerably from model to model . in this section ,
we first describe the evolution of our standard model , and then present the differences in the off - axis shocks and nuclear rings caused by differing @xmath0 and @xmath1 .
the properties of nuclear spirals will be given in section [ sec : nsp ] .
figure [ fig : snap ] plots snapshots of the logarithm of the gas density at a few selected epochs in the inner regions of model cs05bh0 .
the bar is oriented vertically along the @xmath47-axis , and the gas is rotating in the counterclockwise direction relative to the bar .
the images extend to 6 kpc on either side of the center , corresponding to the cr radius , outside of which the gas remains almost unperturbed . ] .
@xcite found that the cr regions exhibit time - dependent flow structures , as reproduced in figure [ fig : comp_ori_new]a .
on the other hand , figure [ fig : snap ] shows that the cr regions in our simulations are quite stable and exhibit only at late time low - amplitude wavelike features entrained by the dense gas located at the bar ends , similar to the results of sph simulations ( e.g. , @xcite ) .
this indicates that the complicated structures seen in @xcite were likely an artifact of the errors in their force transformation .
a striking feature of each of the snapshots shown in figure [ fig : snap ] at times greater than 120 myr is large amplitude oscillations in the density in the ring and dust lanes .
these features are also seen in the simulations of @xcite for spiral shocks , and are attributed to the wiggle instability " . as we will discuss below , this shock instability appears to be caused by vorticity generation in curved shocks .
an introduction of the non - axisymmetric bar potential induces perturbations on the gas orbits , causing them to deviate from circular trajectories .
the gas density increases ( decreases ) in regions where neighboring orbits come close together ( diverge ) . when @xmath75 myr , the overdense regions are preferentially located downstream from the bar major axis ( fig .
[ fig : snap]a ) . at this time
, the overdensity produced by orbit crowding is largest at @xmath76 . since the perturbing force by the bar is weak inside @xmath77 , the overdensity there is correspondingly small and not readily discernible . over time , the overdense regions become narrower and sharper as the bar amplitude grows and eventually develop into shock fronts at around @xmath78 myr . in what follows , we term these narrow shocks off - axis shocks .
a nuclear ring is beginning to shape at this time , as well . to illustrate the formation of nuclear rings in our models , we plot as solid lines in figure [ fig : stream ] instantaneous streamlines of the gas that starts from point a marked at @xmath79 in model cs05bh0 and from @xmath80 in model cs20bh0 with @xmath81 and no bh at @xmath78 myr .
the two dotted circles in figure [ fig : stream]a indicate the inner and outer ilrs at @xmath42 and @xmath82 . note that the thick lines representing the overdense ridges in both models directly cross the outer ilr .
the changes of the azimuthal and radial velocities along the streamlines are shown in figure [ fig : stream]b , c , where the dotted line indicates the equilibrium rotation curve of the model galaxy with no bh .
on emerging from the overdense region ( point a ) , the gas moves radially inward on its epicycle orbit and increases ( decreases ) its azimuthal ( radial ) velocity due to the coriolis force .
it reaches point b closest to the center when it attains @xmath83 and largest @xmath84 .
after this point , it moves radially outward , decreasing @xmath84 until it hits the off - axis shock at point c. the gas loses a significant amount of angular momentum at the shock and begins to fall in .
in addition , the shocked gas is swept by other shocked gas flowing from the bar end regions along the shocks .
note that the shape of the off - axis shocks shown in figure [ fig : stream]a coincides with the gas streamline from points c to d , indicating that all the gas after crossing the shocks moves radially in along the shock fronts in the developing stage of the nuclear rings . as the shocked gas moves along the shock fronts from point c
, it gradually rotates faster again .
when the azimuthal velocity of the gas is increased to the level comparable to the equilibrium circular velocity at some radius @xmath11 , it begins to follow a closed orbit ( point d ) , forming a nuclear ring at that radius .
in other words , the centrifugal barrier inhibits the inflowing gas from moving further in .
regardless of the bh mass , this happens at @xmath85 in models with @xmath74 and @xmath86 in models with @xmath81 ( fig .
[ fig : snap]b , c ) .
the facts that the off - axis shocks penetrate the ilr in bh7/bh8 models , the outer ilr in bh0 models , and that the ring positions are almost independent of @xmath1 when the rings are beginning to form suggest that the ring formation is unlikely to be governed by resonances . for models with low sound speed ,
the shape of nuclear rings is similar to an @xmath2-orbit .
clearly , the presence of the nuclear ring prevents the shocked gas from infalling directly to the nucleus .
since the off - axis shocks are curved , crocco s theorem ensures that vorticity can be generated at the shock fronts .
figure [ fig : pvort ] plots snapshots of the potential vorticity @xmath87 relative to the initial value @xmath88 near the shocks at @xmath89 , 100 , 110 , and 120 myr of the standard model . at @xmath89 myr
, @xmath90 is largest along the shocks .
vorticity produced at the shocks is advected with the background flows and enters the shock fronts at the opposite side after a half revolution .
vorticity grows secularly with time by successive passages across the shocks .
when vorticity achieves substantial amplitudes , it causes the shock fronts to wiggle and fragment into small clumps with high vorticity ( see fig .
[ fig : snap]c ) .
the process of clump formation along the shocks in our models bears remarkable resemblance to the wiggle instability of spiral shocks found by @xcite ( see also @xcite ) .
these clumps are carried radially inward and add to the nuclear ring , making the latter fairly inhomogeneous ( fig .
[ fig : snap]d , e ) .
the off - axis shocks shown in figure [ fig : stream ] are not stationary largely because the bar potential is not fully turned on yet . as the strength of the bar potential keeps increasing , they become stronger and move slowly toward the bar major axis .
gas that is added to the ring from the off - axis shocks has increasingly lower angular momentum with time , causing the ring to shrink in radius with time .
after the bar potential is fully turned on , the off - axis shocks become gradually weaker as the amount of gas in the mid - bar regions lost to the ring increases with time . at the same time , orbital phase mixing and frequent clump collisions in the ring make the latter rounder and align its semimajor axis in the direction perpendicular to the bar major axis .
note that the rings are always attached to the inner end of the off - axis shocks . at @xmath91 myr ,
model cs05bh0 reaches a quasi - steady state in the sense that temporal changes in the overall flow pattern are very slow , and the locations of the shocks and rings do not vary much with time .
figure [ fig : snap]f overplots some of the @xmath92-orbits ( dotted curves aligned vertically ) and @xmath2-orbits ( solid curves aligned horizontally ) , showing that the shape of the nuclear ring matches well with an @xmath2-orbit , while the off - axis shocks closely follow one of the @xmath92-orbits over the whole length of the shocks .
this is because when @xmath74 the impact of thermal pressure on the gas orbits is much smaller than that of the gravitational and centrifugal forces , so that pure orbit theory ( neglecting pressure forces ) is a good description .
when @xmath93 , however , thermal pressure gradients strongly affect gas orbits , and thus the morphology of substructures in the central regions are modified , as we will discuss below .
lcccrrcc cs05bh0 & 0.98 & 5.7 & 12.1 & 10.9 & 47.7 & 2.7 & 7.2 + cs10bh0 & 0.72 & 3.6 & 6.0 & 4.9 & 50.8 & 1.8 & 3.5 + cs15bh0 & 0.55 & 5.7 & 5.7 & 1.0 & 80.4 & 2.3 & 2.6 + cs20bh0 & 0.45 & 5.0 & 3.5 & 1.2 & 71.3 & 1.8 & 1.6 + cs05bh7 & 0.93 & 5.8 & 15.4 & 6.3 & 67.9 & 1.8 & 7.4 + cs10bh7 & 0.63 & 9.8 & 8.1 & @xmath94 & @xmath95 & 1.6 & 4.7 + cs15bh7 & 0.47 & 9.0 & 6.6 & @xmath96 & @xmath97 & 1.7 & 3.0 + cs20bh7 & 0.31 & 5.7 & 2.3 & 1.3 & 59.4 & 1.0 & 1.9 + cs05bh8 & 0.96 & 6.1 & 16.7 & 7.2 & 66.7 & 2.0 & 7.5 + cs10bh8 & 0.74 & 7.4 & 9.6 & @xmath98 & @xmath99 & 2.3 & 6.3 + cs15bh8 & 0.43 & 6.8 & 7.2 & @xmath100 & @xmath101 & 1.4 & 3.0 + cs20bh8 & 0.32 & 21.6 & 5.9 & @xmath102 & @xmath103 & 1.4 & 2.9 even if the gravitational potential is the same , the flow morphology and velocity fields differ considerably depending on the sound speed .
figure [ fig : csvar ] shows instantaneous streamlines plotted over the logarithm of the density distribution in models cs05bh0 and cs20bh0 at @xmath91 myr .
red curves denote the streamlines that go through the off - axis shocks , while those enveloping the off - axis shocks are represented by green curves . in all models ,
the off - axis shocks are almost parallel to @xmath92-orbits .
they start from the bar major axis at the outer ends , offset toward downstream in the mid - bar regions , and connect to the nuclear rings at the inner ends .
the mean offset of off - axis shocks from the bar major axis is larger for models with smaller @xmath0 .
the outer end regions of the off - axis shocks have complicated density structures including the `` 4/1-spiral shocks '' marked with a red arrow in figure [ fig : csvar]a @xcite and the enhanced density ridges ( a white arrow ) termed `` smudges '' by @xcite .
as discussed by @xcite , the 4/1-spiral shocks are produced by collisions of gas moving on @xmath92-orbits with that on the 4/1-resonant family ( e.g. , @xcite ) .
the gas loses angular momentum at the shocks and subsequently switches to lower orbits . as the streamlines in green display , in models with @xmath74 , the 4/1-spiral shocks are quite strong and spatially extended , so that the orbits after the shocks become relatively radial and converge at the opposite side of the bar , building a smudge after about a half revolution .
collisions of streams off the 4/1-spiral shock and the smudge on the same side of the bar funnel the gas at the intersections to an @xmath92-orbit , which are the starting points of the off - axis shocks . when @xmath81 , on the other hand , the 4/1-spiral shocks are short and weak , and the streamlines off the shocks diverge , so that a dense ridge does not form . since the gas becomes less compressible with increasing @xmath0 , steady off - axis shocks in models with large @xmath0 can be supported only in inner regions where the bar perturbations are sufficiently strong .
this explains why the mean offset of the off - axis shocks from the bar major axis becomes smaller as @xmath0 increases ( e.g. , @xcite ) . with weak 4/1-spiral shocks and no smudge ,
the gas in the bar - end regions in model with @xmath81 is comparatively unsteady , sometimes generating small dense blobs that move inward along the off - axis shocks . to quantify the shock properties , we place a slit in each model , indicated by the short white lines in figure [ fig : csvar ] .
the slit starts from @xmath104 and runs perpendicular to the local segment of the off - axis shocks , roughly at @xmath105 .
figure [ fig : slit ] plots the profiles along the slit of surface density , velocities , and the compression factor @xmath106 the last of which can be used as an effective measure of the shock strength ( e.g. , @xcite ) at the shock discontinuities . ] .
the gas is flowing from left to right in the increasing direction of @xmath107 , where @xmath107 measures the distance along the slit from the starting point .
table [ tbl : shock ] gives the shock properties for all models at @xmath91 myr : @xmath108 is the position of the off - axis shocks along the slit , @xmath109 is the peak density after @xmath108 , @xmath110 and @xmath111 are the mach numbers of the incident flows in the directions perpendicular and parallel to the shocks , respectively , @xmath112 is the inclination angle of the preshock velocity relative to the shock front , @xmath113 quantifies the velocity shear in the postshock region , and @xmath114 is the maximum value of the compression factor occurring at the shock front .
it is apparent that the off - axis shocks tend to move toward the bar major axis with increasing @xmath0 , while there is no clear dependence of @xmath108 on the bh mass .
the compression factor at the shock is insensitive to @xmath1 and scales roughly with @xmath0 as @xmath115 .
naively , one would expect that the off - axis shocks become weaker as the sound speed increases , since the density jump in planer isothermal shocks is proportional to @xmath116 . however , this is not the case , as figure [ fig : slit ] and table [ tbl : shock ] demonstrate .
there are several reasons for this .
firstly , it is the velocity component normal to the shock front @xmath117 that determines the shock jump conditions , and because the inclination angle of the streamlines which enter the shock varies with location and with @xmath0 , @xmath117 varies in a complicated fashion .
for example , figure [ fig : slit ] shows that for off - axis shocks formed at @xmath118 , model cs15bh0 with @xmath119 has the largest peak density as well as the largest @xmath120 and @xmath121 . on the other hand ,
model cs10bh0 has the smallest @xmath122 ( with @xmath123 ) and thus the lowest density enhancement . since the sound speed is lower , model cs05bh0 with @xmath122 produces @xmath109 comparable to that in model cs15bh0 .
secondly , we note that the rankin - hugoniot jump conditions for stationary planar shocks are not applicable to the curved and two - dimensional off - axis shocks formed in our simulations . as figure [ fig : csvar ]
displays , the flows are fully two - dimensional in the sense that streamlines diverge before the shocks , and converge after the shock with the radial inflow coming from the regions near the end of the bar . the fact that the compression factor @xmath124 measured at the shock front is smaller than @xmath125 expected from planer isothermal shocks also indicates that the shocks are two dimensional . finally , the density and velocity fluctuations generated by the vortex - generating instability are important around the off - axis shocks especially for models with small @xmath0 , so that the flows are not strictly stationary right after the shocks in model cs10bh0 shown in fig . [
fig : slit ] are due to vortices produced by the instability . ] .
note that the shocked gas has strong velocity shear , amounting to @xmath126 , which is about 10 times larger than the velocity shear arising from galaxy rotation in the solar neighborhood .
such strong shear can stabilize the high - density , off - axis shocks against self - gravity .
lccccc cs05bh0 & 0.63 & 1.22 & 0.92 & 17.2 & 11.9 + cs05bh0 t & 0.62 & 1.17 & 0.89 & 18.4 & 13.1 + cs10bh0 & 0.45 & 0.90 & 0.68 & 30.2 & 17.9 + cs15bh0 & 0.30 & 0.77 & 0.54 & 37.5 & 17.6 + cs20bh0 & 0.23 & 0.71 & 0.49 & 36.8 & 16.9 + cs20bh0 t & 0.20 & 0.73 & 0.48 & 34.7 & 16.1 + cs05bh7 & 0.66 & 1.17 & 0.90 & 18.9 & 13.9 + cs10bh7 & 0.46 & 0.88 & 0.67 & 30.3 & 18.9 + cs15bh7 & 0.28 & 0.80 & 0.52 & 33.9 & 16.9 + cs20bh7 & 0.23 & 0.67 & 0.47 & 41.0 & 19.5 + cs20bh7 t & 0.22 & 0.66 & 0.46 & 43.4 & 19.9 + cs05bh8 & 0.69 & 1.25 & 0.94 & 18.1 & 12.7 + cs10bh8 & 0.47 & 0.89 & 0.67 & 30.9 & 19.4 + cs15bh8 & @xmath127 & @xmath127 & @xmath127 & @xmath127 & @xmath127 + cs20bh8 & @xmath127 & @xmath127 & @xmath127 & @xmath127 & @xmath127 gas that loses angular momentum at the off - axis shocks flows radially inwards and forms a nuclear ring in the central regions . figure [ fig : ringall ] shows diverse morphological features produced in the regions with @xmath128 for all models at @xmath91 myr . figure [ fig : st_ring ] overplots instantaneous streamlines for a few selected models . some models ( with low @xmath0 )
have a nuclear ring together with inner spiral structures , some models ( with high @xmath0 and small @xmath1 ) have a ring with no spirals , while others ( with high @xmath0 and large @xmath1 ) possess only nuclear spirals without an appreciable ring . when @xmath74 , the nuclear rings are quite narrow and clearly decoupled from the nuclear spirals .
even though the ring has a large density , the low sound speed makes the effect of thermal pressure on the gas orbits insignificant .
the gas around the ring in model cs05bh0 thus follows @xmath2-orbits fairly well and the shape of the ring does not deviate considerably from @xmath2-orbits ( fig . [
fig : st_ring]a ) .
when @xmath81 , on the other hand , the pressure force in the central regions becomes important and affects the shape of the gas streamlines .
even the inflowing gas that arrives at the contact points between the off - axis shocks and the nuclear ring takes very different orbits depending on its location .
some gas at the outer parts of the contact points is pushed out by the pressure gradient and follows trajectories that are much rounder than @xmath2-orbits , while the gas in the inner parts is forced to take inner highly eccentric orbits ( fig .
[ fig : st_ring]b ) .
consequently , the gas in the central regions in models cs20bh0 spreads out spatially and forms a ring that is more circular and broader than in model cs05bh0 .
since the presence of a central bh increases the initial angular momentum of the gas in the central regions , the pressure effect becomes less important as @xmath1 increases .
figure [ fig : st_ring ] shows that the pressure distortion of @xmath2-orbits is still significant for @xmath129 , while the gas orbits in the very central parts at @xmath130 remain almost intact when @xmath27 . in model cs20bh8 , some gas at @xmath131 temporarily moves radially outward due to the radial pressure gradient built up by the background gas and is subsequently swept inward by other gas flowing in along the off - axis shocks .
figure [ fig : ring_aden ] plots the radial distribution of the averaged gas surface density @xmath132 , averaged both azimuthally and temporally over @xmath133 myr for all models .
the locations of the ilrs as well as @xmath134 and @xmath135 corresponding to the local maximum and minimum of the @xmath32 curve are indicated by arrows on the abscissa .
table [ tbl : ring ] gives the inner and outer radii , @xmath136 and @xmath137 , of the ring , the mass - weighted ring radius @xmath138 , the peak density @xmath139 , and the mean density @xmath140 of the ring in each model .
here , @xmath136 and @xmath137 are defined as the radii where @xmath141 .
note that models cs15bh8 and cs20bh8 do not harbor a well - defined nuclear ring , as will be discussed in the next section .
all rings that form are located within @xmath142 if there are two ilrs or @xmath143 if there is a single ilr , indicating that the formation of a nuclear ring does not require the presence of two ilrs . however , there is in general no direct connection between the ring positions and @xmath142 or @xmath143 .
when @xmath74 , the rings are all located at @xmath144 , independent of @xmath1 .
the mass - weighted radius is @xmath145 , indicating that the ring position is not governed by the shape of the @xmath32 curve .
when @xmath146 , the ring radius decreases to @xmath147 , again insensitive to the bh mass , consistent with the tendency for the off - axis shocks to move closer to the bar major axis as @xmath0 increases .
rings with @xmath146 have a larger surface density than those with @xmath74 , corresponding approximately to a constant ring mass ( i.e. , @xmath148 ) .
as @xmath0 increases further , high thermal pressure provides strong perturbations for @xmath2-orbits and tend to spread out the gas in the central parts , resulting in a broad distribution of @xmath132 at @xmath149 . when the bh is not massive enough , these perturbations wipe out coherent , weak spiral structures that formed earlier in the nuclear regions . because the presence of a central bh dominates the potential only in the central region , the allowance for the growth of the bh due to mass accretion does not make significant difference in the regions outside the ring .
even inside the ring , figure [ fig : ring_aden ] and table [ tbl : ring ] show that the changes in the ring size @xmath150 and the ring density @xmath151 caused by the temporal change of @xmath1 are less than 4% and 10% , respectively
. we will show below that bh growth in our simulations does not significantly affect the properties of nuclear spirals and mass inflow rates , as well .
high - resolution observations of barred galaxies reveal that some contain nuclear spirals inside a nuclear ring ( e.g. , @xcite ) . as figures [ fig : ringall ] displays ,
some of our models also have spiral structures in their nuclear regions that persist for long periods of time
. other models also have nuclear spirals at early time but they are destroyed as a result of interactions with nuclear rings . in this section ,
we describe the formation and shape of nuclear spirals in detail .
figure [ fig : nsp_class ] schematically summarizes how the type of nuclear spirals changes with time and how long they survive in each model .
table [ tbl : spiral ] lists the properties of nuclear spirals measured at @xmath152 myr for the models that possess long - lasting spirals : @xmath136 and @xmath137 denote the radii of the inner and outer ends of the nuclear spirals , respectively ; @xmath153 and @xmath154 are the azimuthally - averaged and peak surface densities , respectively , at @xmath155 for models cs15bh8 and cs20bh8 and @xmath156 for the other models ; @xmath157 is the pitch angle of the spirals at @xmath158 .
a negative ( positive ) value of @xmath157 indicates leading ( trailing ) spirals .
lccccc cs05bh0 & 0.13 & 0.50 & 0.41 & 7.92 & @xmath159 + cs05bh0 t & 0.20 & 0.45 & 0.56 & 9.84 & @xmath160 + cs05bh7 & 0.02 & 0.45 & 0.92 & 1.95 & @xmath161 + cs10bh7 & 0.05 & 0.42 & 0.16 & 3.86 & @xmath162 + cs05bh8 & 0.02 & 0.55 & 1.00 & 2.07 & @xmath163 + cs10bh8 & 0.02 & 0.40 & 2.28 & 1.91 & @xmath164 + cs15bh8 & 0.02 & @xmath127 & 8.09 & 1.67 & @xmath165 + cs20bh8 & 0.02 & @xmath127 & 170.0 & 1.74 & @xmath166 to study the spiral features , it is convenient to show the logarithm of the gas surface density in logarithmic polar coordinates . figure [ fig : nspbh0 ] plots snapshots of gas surface density of models cs05bh0 and cs10bh0 on the @xmath167 plane .
any coherent features with a positive ( negative ) slope on this plane are leading ( trailing ) waves .
only the regions with @xmath168 are shown .
two horizontal lines mark @xmath169 and @xmath134 where the @xmath32 curve is a locally maximum . at early time , the non - axisymmetric bar potential induces weak @xmath6 perturbations in the central regions .
perturbed gas elements in the galactic plane follow slightly elliptical orbits , which are closed in a frame rotating at @xmath32 and thus precess at a rate @xmath32 near the ilrs when seen in a stationary bar frame . as succinctly depicted in @xcite , due to collisional dissipation of gas kinetic energy occurring on converging orbits , the gas forms spiral structures whose shape depends critically on the sign of @xmath46 such that spirals are leading ( trailing ) where @xmath46 is positive ( negative ) .
figure [ fig : nspbh0 ] indeed shows that when @xmath170 myr the perturbed density is leading at @xmath171 and trailing at @xmath172 , although the trailing features are soon overwhelmed by the nuclear ring . located away from the nuclear ring , however , the inner leading waves are able to grow with time and eventually develop into shock waves at @xmath173 myr .
these nuclear spirals are short , extending over @xmath174 , quite open with a pitch angle of @xmath175 , and almost completely detached from the nuclear ring . figure [ fig : nsprofbh0 ] plots the azimuthal distributions of surface density and velocities of the nuclear spirals at @xmath176 in models cs05bh0 ( solid lines ) and cs05bh0 t ( dotted lines ) when @xmath152 myr . over the course of the orbits , the changes of the radial and azimuthal velocities associated with the spirals amount to @xmath177 , which is indeed large enough to induce shocks .
the peak densities occurring at @xmath178 for model cs05bh0 correspond to shock fronts with a compression factor of @xmath179 .
the density bumps at @xmath180 are produced by waves launched from the inner boundary . in model cs05bh0 t ,
the bh mass is increased to @xmath181 due to mass inflow , which supports slightly stronger , more leading spiral shocks than in model cs05bh0 with no bh . despite continual perturbations by traveling trailing waves propagating from the inner boundary , the nuclear spirals in these models last until the end of the run . that leading nuclear spirals are persistent when the sound speed is small is consistent with the results of sph simulations reported by @xcite . the usual wkb dispersion relation for tightly - wound
, linear - amplitude waves in a non - self - gravitating medium reads @xmath182 where @xmath183 is the wave frequency and @xmath184 and @xmath5 are the radial and azimuthal wavenumbers ( e.g. , @xcite ) . for @xmath6 waves corotating with a bar ( i.e. , @xmath185 ) , equation ( [ eq : disp1 ] ) becomes @xmath186 ( e.g. , @xcite ) , indicating that nuclear spirals corotating with the bar can exist only in the regions between @xmath169 and @xmath142 when there are two ilrs shows that there are low - amplitude waves propagating relative to the bar inside @xmath169 .
such waves can exist inside the inner ilr as long as they satisfy equation ( [ eq : disp1 ] ) in the wkb limit . ] .
however , the top row in figure [ fig : nspbh0 ] reveals that the nuclear spirals in model cs05bh0 extend slightly inward of @xmath169 .
this seemingly contradicting result is due to nonlinear effects which are not captured by wkb theory .
the velocity perturbations associated with these spirals are so large that fluid elements just outside @xmath169 can move across the inner ilr over the course of their epicycle orbits , providing perturbations for the gas at @xmath187 that responds passively . in model cs05bh0 ,
the radial velocity perturbation is @xmath188 .
since the epicycle frequency is @xmath189 at @xmath190 , the corresponding radial amplitude of the epicycle orbits is estimated to be @xmath191 , which is in good agreement with the radial extent of the nuclear spirals inward of @xmath169 .
models without a bh and with @xmath192 do form nuclear spirals at early time , but they are all transient , lasting less than 200 myr .
the bottom row of figure [ fig : ringall ] shows how nuclear spirals are destroyed in model cs10bh0 .
the nuclear ring in this model is not only located inside @xmath134 but also has large thermal pressure , continuously generating sonic perturbations that propagate radially inward . because of the background shear , the perturbations are preferentially in the form of trailing waves which interact destructively with the leading spirals that formed at @xmath171 , destroying the latter .
the destruction of nuclear spirals happens at @xmath193 myr for model cs10bh0 .
this occurs earlier as @xmath0 increases , since disturbing pressure perturbations are correspondingly stronger . since the presence of a central bh greatly changes the @xmath32 curve in the central regions , it is interesting to explore how the morphologies of nuclear spirals depend on the bh mass . in bh7 and bh8 models with a single ilr ,
stationary waves in the bar frame can exist inside @xmath143 all the way down to the center .
figure [ fig : nspbh7 ] plots snapshots of gas surface density for models cs05bh7 and cs10bh7 on the logarithmic polar plane . the positions of @xmath134 and @xmath135 are indicated by the horizontal lines .
as expected , the overdense perturbations produced by orbit crowding at early time ( @xmath170 myr ) have leading configurations at @xmath194 and trailing configurations at @xmath195 or @xmath172 .
the overdense regions at @xmath172 are subsequently wiped out as the nuclear ring forms , while those at @xmath171 grow into trailing nuclear spirals . in model cs05bh7 ,
the leading spirals at @xmath194 also grow slightly until @xmath196 myr to temporarily form `` inner - trailing and outer - leading '' structures represented by the double cross - hatching in figure [ fig : nsp_class ] . as trailing perturbations from both the inner trailing spirals and the outer nuclear ring propagate and interfere with the leading spirals , it becomes increasingly difficult to identify coherent spiral structures at @xmath194 .
on the other hand , the trailing spirals at @xmath195 keep growing until @xmath197 myr after which their amplitude of @xmath198 remains more or less constant .
they are approximately logarithmic in shape with a pitch angle of @xmath199 . unlike in model cs05bh0 where leading spirals are actually shocks , the trailing nuclear spirals in models cs05bh7 are relatively weak ( with the maximum compression factor of @xmath200 at @xmath152 myr ) , and never develop into shocks . in model cs10bh7 ( bottom row of fig .
[ fig : nspbh7 ] ) , the nuclear spirals have larger @xmath184 and thus are more open than those in the @xmath74 counterparts ( see , e.g. , eq . [ [ eq : disp2 ] ] ) . since the nuclear ring in this model forms at @xmath201
, it can directly destroy the leading spiral at @xmath194 , and feeds the inner trailing spirals by supplying trailing perturbations .
thus , the inner spirals grow both in strength and spatial extent to make contact with the nuclear ring at @xmath202 myr . at this time
, all parts of the nuclear spirals turn to shocks with the maximum density of @xmath203 and the corresponding compression factor of @xmath204 at @xmath205 .
gas passing through the spiral shocks loses angular momentum , increasing the mass inflow rate at the inner boundary . as the amount of gas lost in the nuclear regions increases , the density of the nuclear spirals decreases with time , but they remain as shocks with the compression factor of @xmath206 until the end of the run . in models
cs15bh7 and cs20bh7 , nuclear spirals start out with inner - trailing and outer - leading shapes , and evolve into trailing - only configurations , as in the other bh7 models .
however , the highly eccentric orbits of the gas affected by thermal pressure dismantle the inner spiral structures almost completely in these models : thermal perturbations are so strong that a central bh with @xmath129 can not enforce circular orbits in the very nuclear regions ( see fig . [
fig : st_ring]c ) .
since the @xmath32 curve decreases monotonically with @xmath36 in bh8 models , nuclear spirals , if they exist , are all trailing as evident in figure [ fig : nspbh8 ] . for model cs05bh8 ( top row of fig .
[ fig : nspbh8 ] ) , the spirals evolve almost independently of , and are well separated from , the nuclear ring . at @xmath152 myr , they have a very small pitch angle @xmath207 corresponding to @xmath208 owing to the strong background shear .
they are also very weak , with an amplitude of @xmath209 at @xmath210 .
when @xmath146 , the pitch angle is increased to @xmath211 and the nuclear spirals extend outward all the way to the nuclear ring . except near the contact points ,
the spirals are still decoupled from the ring . with quite a large value of @xmath212 ,
the component of the rotational velocity perpendicular to the spirals is not large enough to induce shocks .
we have seen earlier that the large thermal pressure in the rings of models with @xmath213 provides strong perturbations that make the gas orbits near the galaxy center highly eccentric , destroying nuclear spirals in bh0 and bh7 models . in bh8 models , however , the situation is quite different since a central bh with @xmath27 dominates the potential , keeping the orbits almost circular in the very central regions .
even with a large thermal pressure , the gas orbits there can not be very eccentric , so that the nuclear spirals are protected from disruptive pressure perturbations .
on the other hand , the ring material is quite distributed because of the large pressure gradients , feeding a trailing nuclear spiral that grows strongly in models cs15bh8 and cs20bh8 . at @xmath214 myr ,
the spirals in models cs15bh8 and cs20bh8 turn into shocks and touch the densest parts of the ring located at @xmath215 .
because of the larger pitch angles , the shocks are stronger in the outer parts ; the portions at @xmath31 are weak shocks with density jumps of only @xmath216 . similarly , the shocks in model cs20bh8 are stronger than in model cs15bh8 since the former has more open spirals and a denser ring . in both models ,
the shocks near the ring are so strong that even the gas constituting the ring suffers from a significant loss of angular momentum at the intersecting points . at @xmath217 myr , the ring material is essentially dissected by the trailing spiral shocks and gradually moves toward the center . as the ring material continues to flow in , the spirals appear as a direct continuation of the off - axis shocks ( @xmath218 myr ) , consistent with the results of @xcite .
figure [ fig : nsprofbh8 ] plots the radial distributions of the maximum and mean densities as well as the maximum compression factor in the inner @xmath219 regions of models cs15bh8 and cs20bh8 at @xmath152 myr .
the inflow of the ring material in model cs20bh8 is quite strong that the gas is collected in the nuclear regions with @xmath220 ; the amount of the gas that goes out through the inner boundary is much smaller than that comes in . with weaker shocks and thus less angular momentum loss ,
on the other hand , the destroyed ring gas in model cs15bh8 is still mostly at @xmath221 at this time .
note that the density enhancement at @xmath222 resulting from the dissolution of the ring is @xmath223 and @xmath224 in models cs15bh8 and cs20bh8 , respectively : the mean contrast of the spirals that are weak shocks with the compression factor @xmath225 is @xmath226 at @xmath222 in both models .
this increase of the gas surface density in the central regions is the primary reason for enhanced mass inflow rates at late time in bh8 models with large @xmath0 .
galactic bars are considered to be a promising means of transporting gas to the centers of galaxies to fuel supermassive bhs and produce agns . since
our numerical models use a cylindrical grid with a circular boundary , they are ideally suited to study how the mass accretion rate depends on the gas sound speed and the bh mass .
we assume that all the gas that crosses the inner boundary located at @xmath227 in our models is accreted to the central bh . in reality ,
a large amount of mass in gas near the bh may change the gas orbits by providing pressure and gravitational forces .
therefore , @xmath26 through the inner boundary that we measure is likely to be an upper limit to the real mass accretion rate to the bh .
in addition , @xmath26 is likely to depend on the inner boundary size especially when gas orbits are highly eccentric near the inner boundary .
note that in non - self - gravitating , isothermal , and unmagnetized systems , @xmath26 resulting from simulations is linearly proportional to the adopted initial surface density @xmath228 ; all the models presented here take @xmath68 .
lcc cs05bh0 & @xmath229 + cs05bh0 t & @xmath230 + cs10bh0 & @xmath231 + cs15bh0 & @xmath232 + cs20bh0 & @xmath233 + cs20bh0 t & @xmath234 + cs05bh7 & @xmath235 + cs10bh7 & @xmath236 + cs15bh7 & @xmath237 + cs20bh7 & @xmath238 + cs20bh7 t & @xmath239 + cs05bh8 & @xmath240 + cs10bh8 & @xmath241 + cs15bh8 & @xmath242 + cs20bh8 & @xmath243 in general , @xmath26 would be large if the gas orbits near the center are highly eccentric or radial , while circular orbits would make @xmath26 quite small .
this expectation is consistent with figure [ fig : mdot ] which plots the temporal evolution of the mass inflow rates for all models .
the corresponding accreted gas mass @xmath244 over @xmath245 myr is given in table [ tbl : macc ] .
clearly , @xmath26 is larger for models with larger @xmath0 and no bh , compared to models with a massive bh of @xmath27 .
when @xmath74 , the nuclear spirals are well separated from the nuclear rings , and the departure of the gas orbits from a circular shape near the inner boundary is small , resulting in quite small values of @xmath26 ( @xmath246 ) . in bh8 models ,
the presence of a central bh makes @xmath26 smaller by about an order of magnitude than in bh0 models by providing strong axisymmetric gravitational potential near the center . in bh7 models ,
the bh potential is not strong enough to circularize the eccentric orbits , giving rise to @xmath26 only slightly smaller than that in bh0 models .
increasing @xmath0 enhances @xmath26 because pressure perturbations become stronger and the nuclear ring tends to be located closer to the center , both of which strongly affect the gas orbits in the central region . when @xmath146 , @xmath26 is increased by about an order of magnitude compared to the cases with @xmath74 , but is still less than @xmath247 except for a brief time interval around @xmath91 myr in model cs100bh7 when the nuclear spirals shock the central gas and cause enhanced inflows .
when @xmath248 , the gas orbits in bh0 and bh7 models are quite eccentric near the center , so that some gas with highly radial orbits plunges directly into the inner boundary , increasing @xmath26 dramatically compared to the lower @xmath0 counterparts .
the associated gas mass accreted to the galaxy center is of order of @xmath249 over 500 myr , suggesting that a strong galactic bar like studied in this work can be an appealing means for the growth of supermassive bhs provided the gaseous medium has a large ( effective ) sound speed ( cf .
@xcite ; see also , e.g. , @xcite for review ) . in bh8 models , on the other hand , the gas orbits in the central parts are not perturbed much ( since the initial angular momentum is overwhelmingly large ) . as the nuclear spirals grow into shocks , however , the nuclear gas as well as the ring material in these models drift inward slowly , increasing @xmath26 over time . in models with @xmath27 , the late - time values of @xmath26 in models with
@xmath250 or @xmath22 are larger than @xmath3 , sufficient to power agns in seyfert galaxies ( e.g. , @xcite ) .
finally , we present the mass inflow rates resulting from the models in which @xmath1 is varied self - consistently with @xmath26 .
the left panels of figure [ fig : mdot_mbht ] compare @xmath26 from models cs05bh0 t and cs20bh0 t with those from the fixed-@xmath1 counterparts , while the bottom panels plot the temporal evolution of the bh mass in the former models . in model cs05bh0 t
, the total gas mass accreted over 500 myr is @xmath251 , with the corresponding increment of the equilibrium rotational velocity of @xmath252 at the inner boundary .
since this is three times smaller than the initial circular velocity there , the bh mass does not affect @xmath26 much , as figure [ fig : mdot_mbht ] illustrates .
in the case of model cs20bh0
t , @xmath1 attains @xmath253 at @xmath152 myr , which is large enough to be dynamically important . when @xmath81 , however , the gas orbits are highly eccentric and @xmath26 is insensitive to the bh mass as long as @xmath254 ( see fig .
[ fig : mdot]d ) . as
the rignt panels of figure [ fig : mdot_mbht ] show , the increase of @xmath1 over 500 myr in model cs20bh7 t is less than a factor of two , corresponding to the equilibrium circular velocity 1.3 times the initial value . with an enhanced centrifugal barrier
, the resulting @xmath26 becomes gradually smaller than the case with fixed @xmath1 , but by less than , on average , @xmath255 in @xmath91-@xmath245 myr .
therefore , we conclude that the effect of bh growth due to gas accretion on the mass inflow rate as well as bar substructures is not significant for the models we have considered .
we have presented detailed numerical models that explore the formation of substructures produced by the gas flow in barred galaxies .
previous models based on particle simulations ( e.g. , @xcite ) did not have sufficient resolution to resolve nuclear spirals . on the other hand ,
studies that used the grid - based code cmhog ( e.g. , @xcite ) unknowingly made mistakes in the force evaluation for the bar potential , so that the results needed to be recomputed . in this paper
we have corrected the errors in the original cmhog code and run high - resolution hydrodynamical simulations . to resolve the nuclear regions ,
we employ a logarithmically - spaced cylindrical grid , with a zone size of @xmath256 at @xmath168 where nuclear rings and spirals form .
we have included the potential from a central bh , and studied the flow properties as the mass of the bh @xmath1 and the sound speed @xmath0 in the gas are varied . for simplicity , the effects of gaseous self - gravity and magnetic fields are not included .
the main results of the present work are summarized as follows : \1 .
_ off - axis shocks _ the imposed non - axisymmetric bar potential provides gravitational torques to the otherwise circular - rotating gas , perturbing its orbit . the perturbed orbits crowd at the downstream sides of the bar major axis and produce overdense ridges that eventually develop into off - axis shocks . at a quasi - steady state ,
the off - axis shocks are overall almost parallel to @xmath92-orbits : they start from the bar major axis at the outer ends , are gradually displaced downstream as they move inward , and connect to the nuclear ring at the inner ends . while the positions of the off - axis shocks are almost independent of @xmath1 , since the effect of a bh is negligible at large radii , they depend on @xmath0 in such a way that the shocks are , on average , located closer to the bar major axis as @xmath0 increases .
this is primarily because gas with larger @xmath0 should be more strongly perturbed to induce shocks , which occurs deeper in the potential well and thus results in shocks on lower @xmath92-orbits @xcite .
the off - axis shocks are in general curved .
flow streamlines are complicated near the shocks in that they diverge before the shocks and are promptly swept inwards by inflowing gas right after the shocks .
therefore , the usual ranking - hugoniot jump conditions for planar , one - dimensional shocks are not applicable to the off - axis shocks .
in fact , the shock strength , as measured by the peak density @xmath109 at @xmath105 from the center , is @xmath257 for models with no bh and do not sensitively depend on @xmath0 .
the compression factor of the off - axis shocks are insensitive to the bh mass and depends on @xmath0 roughly as @xmath115 .
the off - axis shocks have very strong velocity shear amounting to @xmath258 .
this strong shear may suppress the growth of gravitational instability in the high - density off - axis shocks when self - gravity is included .
_ nuclear rings _
when gas passes through the off - axis shocks , it loses angular momentum , flows inward , and forms a nuclear ring where the centrifugal force balances the external gravity .
the ring is attached to the inner ends of the off - axis shocks and thus becomes smaller in size as @xmath0 increases .
the rings that form in our models are all located inside the ( outer ) ilr , but this does not imply that the ring formation is related to the ilrs .
when @xmath0 is small , the pressure perturbations on the gas orbits are so weak that the rings are quite narrow and their shape is well described by @xmath2-orbits .
the mean radius is @xmath259 when @xmath74 and @xmath260 when @xmath146 , independent of the bh mass .
this suggests that the ring position is not determined by the @xmath32 curve , hence by the ilrs .
when @xmath248 , on the other hand , large thermal pressure strongly affects the gas orbits in the nuclear rings .
for example , some gas near the contact points between the ring and the off - axis shocks is forced out to follow relatively round orbits , while other gas is pulled in radially to make very eccentric orbits .
these diverse gas orbits near the contact points tend to spread out the ring material , making the rings much broader than in models with smaller @xmath0 .
_ nuclear spirals _
since even the non - axisymmetric bar potential is nearly axisymmetric in the central parts , the gaseous responses inside a nuclear ring are not as dramatic as in off - axis shocks . nevertheless , non - axisymmetric @xmath6 perturbations are able to grow inside a ring and develop into nuclear spirals that persist for a long period of time , provided that @xmath0 is small or @xmath1 is large .
although all models have weak spiral structures at early time , in models with large @xmath0 they are soon destroyed by eccentric gas orbits as well as perturbations induced by the large pressure in the rings unless the bh mass is very large .
when @xmath27 , the disruptive pressure perturbations from the rings can not penetrate the very central parts where the gas has extremely large initial angular momentum . in this case , the nuclear spirals are protected from the surrounding and thus are long - lived .
the shape of nuclear spirals is determined by the sign of @xmath46 such that spirals that form in the regions where @xmath46 is positive ( negative ) are leading ( trailing ) , confirming the theoretical expectations of @xcite . with no bh , only the model with @xmath74 has persistent leading spirals in the regions where the @xmath32 curve is an increasing function of @xmath11 .
the leading spirals in this model are quite strong and develop into shocks with the peak density @xmath261 and the compression factor @xmath262 at @xmath176 at the end of the run .
models with @xmath129 initially have hybrid features comprising of trailing spirals at @xmath195 and leading spirals at @xmath194 , where @xmath135 and @xmath134 refer to the radii where @xmath32 curve attains a local minimum and maximum , respectively .
when @xmath263 , however , the leading parts in the hybrid spirals are destroyed by the trailing pressure waves launched by the ring , leaving only the weak trailing spirals behind . when @xmath213 , both leading and trailing spirals are destructed completely by the pressure perturbations .
in models with @xmath27 , @xmath264 in the whole nuclear regions , so that nuclear rings are always trailing . when @xmath263 , the spirals well separated from the ring are weak and tightly wound .
when @xmath213 , on the other hand , the trailing spirals are fed by the gas in the nuclear ring and grow into shocks , the outer ends of which join the inner ends of the off - axis shocks smoothly .
although the nuclear spirals in these models have only modest density contrasts of @xmath265 at @xmath222 , the background density resulting from the inflows of the ring material is greatly enhanced ( by more than two orders of magnitude when @xmath81 ) .
\4 . _ mass inflow rates _
although gas experiences a significant loss in angular momentum when it meets off - axis shocks during galaxy rotation , this does not necessarily translate into the mass inflow all the way to the galaxy center . in models with @xmath263 , a narrow nuclear ring formed by gas with @xmath2-orbits
inhibits further inflows of the gas , resulting in the mass inflow rate @xmath26 through the inner boundary less than @xmath3 , regardless of the bh mass .
the mass inflow rates are greatly enhanced to @xmath266 in models with @xmath267 and @xmath213 or @xmath27 and @xmath81 for different reasons depending on the bh mass .
when @xmath268 , the gas orbits are affected by thermal pressure and some gas in the ring can take highly eccentric orbits , directly falling into the inner boundary . in models with @xmath269 , on the other hand ,
the gas orbits near the center are more or less circular , but the density in the nuclear regions is enhanced greatly because of strong nuclear spirals , increasing @xmath26 .
nuclear rings play an important role in evolution of barred galaxies by providing sites of active star formation near the centers ( e.g. , @xcite ) .
rings certainly consist of gas that migrates from outer parts inward by losing angular momentum at off - axis shocks , but what stops further migration to form a ring remains a matter of debate . as a trapping mechanism of the ring material , @xcite proposed the non - axisymmetric bar toque that forces gas to accumulate between two ilrs or at a single ilr depending on the shape of the gravitational potential ( see also @xcite ) , while @xcite favored the gas transitions from @xmath92- to @xmath2-orbits rather than orbital resonances
however , our numerical results show that a ring is formed at early time because of the centrifugal barrier that the migrating material feels . later on ,
nuclear rings slowly shrink in size as gas with lower angular momentum gas is continuously added .
although nuclear rings are located in between two ilrs in models with no bh , this is a coincidence .
models with a central bh show that the specific ring positions are insensitive to the shape of the @xmath32 curves , suggesting that nuclear ring formation is not a consequence of the orbital resonances .
in fact , the gas flows that produce the ring are not in force balance and have a large radial velocity , so that the concept of resonances and ilrs is not applicable to nuclear rings ( e.g. , @xcite ) .
in addition , the notion of @xmath2-orbits as the gas trapping locations is meaningful only for small @xmath0 .. ] when the sound speed is large , thermal pressure at the contact points between the off - axis shocks and nuclear ring causes the gas orbits to deviate from @xmath2-orbits considerably .
the results of our simulations suggest that not all barred - galaxies possess nuclear spirals at their centers .
long - lasing nuclear spirals exist only when either the sound speed is small or the bh mass is large : they do not survive in models with _ both _
large @xmath0 _ and _ small @xmath1 .
two common views regarding the nature of nuclear spirals are low - amplitude density waves and strong gaseous shocks ( see e.g. , @xcite ) . and , our simulations indeed show that they are either tightly - wound density waves or shocks when the pith angle is relatively large ( @xmath270 ) .
one may speculate that nuclear spirals are more likely to be shocks rather than density waves when @xmath0 is small .
in contrast to this prediction , however , nuclear spirals in models with @xmath27 are density waves when @xmath0 is small and shocks when @xmath0 is large .
this is of course because as @xmath0 increases , waves in nuclear regions tend to be more open ( with smaller @xmath271 ) in the beginning , and they are subsequently supplied with more gas from the rings as they grow .
this is entirely consistent with the results of @xcite who used sph simulations to show that nuclear spirals are supported by shocks when @xmath93 in models with a massive bh .
we note however that weak trailing spirals seen in our model cs10bh8 are absent in model m2 ( @xmath146 and @xmath27 ) of @xcite , which is presumably due to an insufficient number of particles to resolve nuclear spirals in their sph simulations .
of 12 models with differing @xmath0 and @xmath21 that we have considered , only 1 model possesses leading spirals , suggesting that galaxies with leading nuclear spirals would be very uncommon in nature . to our knowledge , only two galaxies , ngc 1241 and ngc 6902 , are known to possess leading features in the nuclear regions @xcite emissions tracing young stars , while those in ngc 6902 are observed in the k@xmath272 band tracing old populations .
it is uncertain how these stellar features are related to gaseous nuclear spirals studied in this paper . ] .
based on our simulations , the existence of leading spirals at centers requires two stringent conditions : ( 1 ) the gas should be dynamically cold enough to protect nuclear spirals from nuclear rings and ( 2 ) there should be a wide range of radii with @xmath264 in the central parts , which can be easily accomplished when there are two ilrs ( or without a strong central mass concentration ) .
the second condition is consistent with the linear theory that predicts short leading waves propagating outward from the inner ilr @xcite .
the facts that ngc 6902 is a barred - spiral galaxy @xcite and does not show significant @xmath273-ray emissions indicative of agn activities @xcite are not inconsistent with the second requirement for the existence of leading nuclear spirals . since ngc 1241 is a seyfert 2 galaxy with an estimated bh mass of @xmath274 @xcite , however , the nuclear star - forming regions in this galaxy are unlikely to be associated with gaseous nuclear spirals studied in this work . finally , we discuss the mass inflow rates derived in our models in regard to powering agns in seyfert galaxies .
the mass accretion rate is often measured by the eddington ratio defined by @xmath275 , where @xmath276 and @xmath277 denote the bolometric and eddington luminosities of an agn and @xmath278 is the mass - to - energy conversion efficiency of the accreted material .
observations indicate that @xmath279 for classical seyfert 1 galaxies with broad iron emission lines ( e.g. , @xcite ; see also @xcite ) , while @xmath280 for low - luminosity seyfert 1 agns ( e.g. , @xcite ) . in our numerical models , the mass inflow rates are larger for models with smaller @xmath1 and larger @xmath0 .
taking @xmath281 ( e.g. , @xcite ) , the mass inflow rates amount to @xmath282 for @xmath27 and @xmath283 , @xmath284 for @xmath27 and @xmath81 or for @xmath129 and @xmath146 , and @xmath285 for @xmath129 and @xmath93 . for classical seyfert galaxies with strong bars
, this suggests that the masses of central bhs are likely to be less than @xmath286 , which appears to be consistent with the measured values from the relation between the bh masses and the velocity dispersions of stellar bulges ( e.g. , @xcite ) and reverberation mapping techniques ( e.g. , @xcite ) . of course
, this result may depend on many factors such as the axis ratio and strength of the bar , presence of self - gravity and magnetic fields , gas cooling and heating , turbulence , etc .
, all of which would affect gas dynamic significantly .
extending the present work to include these physical ingredients would be an important direction of future research .
we gratefully acknowledge helpful discussions with h. ann , j. goodman , m. g. lee , e. ostriker , d. richstone , s.tremaine , & j .- h .
woo , especially f. combes on the shapes of nuclear spirals and bar torque .
we are also grateful to w. maciejewski for stimulating comments on the manuscript and to the referee for a thoughtful report .
this work was supported by the national research foundation of korea ( nrf ) grant funded by the korean government ( mest ) , no .
2010 - 0000712 .
in this appendix we describe the gravitational potential of the model galaxy that maintains the rotation of the ( non - self - gravitating ) gas disk .
the potential is comprised of four component : stellar disk , bulge , bar , and bh . for the disk ,
we take a kuzmin - toomre model with surface density @xmath287 where @xmath288 and @xmath289 are constants @xcite .
the corresponding gravitational potential at the disk midplane is @xmath290 @xcite .
we take @xmath291 and @xmath292 in our simulations . and
a surface density profile that is singular at @xmath293 ( see their eq . ( 1 ) ) .
our adopted parameters nevertheless give the rotational curves similar to those shown in their figure 1 . ]
the total disk mass is @xmath294 . for the bulge , we use a modified hubble profile with volume density @xmath295 and the potential @xmath296 where @xmath71 and @xmath297 are the central density and the characteristic size of the bulge , respectively .
we take @xmath298 and @xmath299 , with the corresponding bulge mass of @xmath300 within @xmath301 .
the bar is modeled by a @xcite ellipsoid with volume density @xmath302 where @xmath70 is the bar central density , @xmath303 with @xmath304 being the cartesian coordinates , and @xmath305 and @xmath306 are the semimajor and semiminor axes of the bar , respectively .
the exponent @xmath307 measures the central concentration of the bar density distribution . in this work ,
we fix the bar parameters to @xmath308 , @xmath309 , @xmath310 , and @xmath311 for all models .
the total mass of the bar is then @xmath312 and the bar quadrupole moment is @xmath313/(5 + 2n ) = 4.5\times 10^{10}{\ ; { \rm m}_{\odot}}{\;{\rm kpc}}^2 $ ] , corresponding to a strong bar . when @xmath308 , the bar potential is given explicitly as @xmath314,\end{aligned}\ ] ] where the coefficients @xmath315 s are defined in @xcite . as the bar pattern speed , we choose @xmath20 which places the corotation resonance at @xmath39 .
finally , for a central bh with mass @xmath1 , we use a plummer potential @xmath316 with the softening radius @xmath317 pc .
we take @xmath318 , @xmath319 , and @xmath320 to study the effects of the central mass concentration on the bar substructures . with @xmath321 , @xmath322 , @xmath323 ,
the bh affects the rotation curve only in the very central regions ( with @xmath324 ) .
figures [ fig : rotcurve ] and [ fig : angfreq ] plot the rotational velocity for models with @xmath325 and the angular frequency curves for all models , respectively .
ann , h. b. , & thakur , p. 2005
, , 620 , 197 athanassoula , e. 1992a , , 259 , 328 athanassoula , e. 1992a , , 259 , 345 agertz o. et al . , 2007 , , 380 , 963 barth , j. , ho , c. , & filippenko , v. 1995 , , 110 , 3 bian , w. , & gu , q. 2007 , , 657 , 159 binney , j. , & tremaine , s. 2008 , galactic dynamics , 2nd ed . , ( princeton university press ) buta , r. 1986 , , 61 , 609 buta , r. , & combes , f. 1996 , fund . cosmic phys . , 17 , 95 colella , p. , & woodward , p. r. 1984 , j. comput . phys . , 54 , 174 combes , f. 1996 , barred galaxies , eds . r. buta , d. a. crocker , & b. g. elmegreen , aps conference , 91 , 286 contopoulos , g. , & grosbl , p. 1989
, , 1 , 261 daz , r. j. , dottori , h. , vera - villamizar , n. , & carranza , g. 2003 , , 597 , 860 denney , k. d. , peterson , b. m. et al .
2010 , , 721 , 715 desroches , l .- b . ,
& ho , l. c. 2009 , , 690 , 267 englmaier , p. , & gerhard , o. 1997 , , 287 , 57 englmaier , p. , & shlosman , i. 2000 , , 528 , 677 ferrers , n. m. 1887 , q.j.pure appl .
, 14 , 1 fabian , a. c. , vasudevan , r. v. , & gandhi , p. 2008
, , 385 , l43 field , g. b. , goldsmith , d. w. , & habing , h. j. 1969 , , 155 , l149 friedli , d. , & benz , w. 1993 , , 268 , 65 garcia - barreto , j. a. , downes , d. , combes , f. , gerin , m. , magri , c. , carrasco , l. , & cruz - gonzalez , i. 1991 , , 244 , 257 goldreich , p. , & tremaine , s. 1978 , , 222 , 850 grosbl , p. 2003
, lecture notes in physics , vol . 626 , galaxies and chaos , eds .
g. contopoulos , n. voglis ( springer : berline ) , p201 gltekin , k. , richstone , d. o. , gebhardt , k. , lauer , t. r. , tremaine , s. , aller , m. c. , bender , r. , dressler , a. , faber , s. m. , filippenko , a. v. , green , r. , ho , luis c. , kormendy , j. , magorrian , j. , pinkney , j. , siopis , c. 2009 , , 698 , 198 heller , c. h. , & shlosman , i. 1994 , , 424 , 84 ho , l .
c. 2008 , , 46 , 475 knapen , j. h. , shlosman , i. , & peletier , r. f. 2000 , , 529 , 93 kim , j. , ryu , d. , jones , t. w. , hong , s. s. 1999 , , 514 , 506 kim , w .- t . , & ostriker , e. c. 2006 , , 646 , 213 kuzmin , g. 1956 , astron .
zh , 33 , 27 laurikainen , e. , salo , h. , buta , r. , & vasylyev , s. 2004 , , 355 , 1251 maciejewski , w. , teuben , p. j. , sparke , l. s. , & stone , j. m. 2002 , , 329 , 502 maciejewski , w. 2004a , , 354 , 883 maciejewski , w. 2004b , , 354 , 892 maoz , d. , barth , j. , ho , c. , sternberg , a. , & filippenko , v. 2001 , , 121 , 3048 martini , p. , regan , m. r. , mulchaey , j. s. , & pogge , r. w. 2003a , , 146 , 353 martini , p. , regan , m. r. , mulchaey , j. s. , & pogge , r. w. 2003b , , 589 , 774 martinez - valpuesta , i. , shlosman , i. , & heller , c. 2006 , , 637 , 214 mazzuca , l. m. , knapen , j. h. , veilleux , s. , & regan , m. w. 2008 , , 174 , 337 mckee , c. f. , & ostriker , j. p. 1977
, , 218 , 148 mckee , c. f. , & ostriker , e. c. 2007 , , 45 , 565 meyer - hofmeister , e. , & meyer , f. 2011 , , 527 , a127 patsis , p. a. , & athanassoula , e. 2000 , , 358 , 45 peterson , b. m. 1997 , an introduction to active galactic nuclei ( cambridge univ . press : cambridge ) pfenniger , d. 1984 , , 134 , 373 piner , b. g. , stone , j. m. , & teuben , p. j. 1995 , , 449 , 508 price d. j. 2008 , j. comput . phys . , 227 , 10040 prieto , m. a. , maciejewski , w. , & reunanen , j. 2005 , , 130 , 1472 read , j. i. , hayfield , t. , & agertz , o. 2010 , , 405 , 1513 regan , m. w. , & mulchaey , j. s. , 1999 , , 117 , 2676 regan , m. w. , & teuben , p. j. 2003 , , 582 , 723 regan , m. w. , & teuben , p. j. 2004 , , 600 , 595 roberts , w. , huntley , m. , & van albada , g. d. 1979 , , 233 , 67 sanders , r. h. , & huntley , j. m. 1976 , , 209 , 53 schwarz , m. p. 1981
, , 247 , 77 shlosman , i. , frank , j. , & begelman , m. c. 1989 , nature , 338 , 45 shlosman , i. , begelman , m. c. , & frank , j. 1990 , nature , 345 , 679 springel , v. , yoshida , n. , & white , s. d. m. 2001 , new astronomy , 6 , 79 thakur , p. , ann , h. b. , & jiang , i. 2009 , , 693 , 586 toomre , a. 1963 , , 138 , 385 van albada , g. d. , & roberts , w. w. 1981 , , 246 , 740 van de ven , g. , & fathi , k. 2010 , , 723 , 767 volonteri , m. 2010 , , 18 , 279 wada , k. , & koda , j. 2004 , , 349 , 270 watabe , y. , kawakatu , n. , & imanishi , m. 2008 , , 677 , 895 yu , q. , & tremaine , s. 2002 , , 335 , 965 yuan , c. , & yen .
d. c. c. 2005 , j. korean astron . | the inner regions of barred galaxies contain substructures such as off - axis shocks , nuclear rings , and nuclear spirals .
these substructure may affect star formation , and control the activity of a central black hole ( bh ) by determining the mass inflow rate .
we investigate the formation and properties of such substructures using high - resolution , grid - based hydrodynamic simulations .
the gaseous medium is assumed to be infinitesimally - thin , isothermal , and non - self - gravitating .
the stars and dark matter are represented by a static gravitational potential with four components : a stellar disk , the bulge , a central bh , and the bar . to investigate various galactic environments , we vary the gas sound speed @xmath0 as well as the mass of the central bh @xmath1 .
once the flow has reached a quasi - steady state , off - axis shocks tend to move closer to the bar major axis as @xmath0 increases .
nuclear rings shrink in size with increasing @xmath0 , but are independent of @xmath1 , suggesting that ring position is not determined by the lindblad resonances .
rings in low-@xmath0 models are narrow since they are occupied largely by gas on @xmath2-orbits and well decoupled from nuclear spirals , while they become broad because of large thermal perturbations in high-@xmath0 models .
nuclear spirals persist only when either @xmath0 is small or @xmath1 is large ; they would otherwise be destroyed completely by the ring material on eccentric orbits .
the shape and strength of nuclear spirals depend sensitively on @xmath0 and @xmath1 such that they are leading if both @xmath0 and @xmath1 are small , weak trailing if @xmath0 is small and @xmath1 is large , and strong trailing if both @xmath0 and @xmath1 are large . while the mass inflow rate toward the nucleus is quite small in low-@xmath0 models because of the presence of a narrow nuclear ring , it becomes larger than @xmath3 when @xmath0 is large , providing a potential explanation of nuclear activity in seyfert galaxies . |
a feshbach resonance can occur in collision events , when a closed channel bound state of a specific electronic level of an atom is coupled to the scattering continuum of an open channel of another electronic level of the atom @xcite .
if the energy of the discrete closed channel state is very close to that of the scattering particles , the cross section is resonantly enhanced , due to the temporary capture of the particles in the quasi - bound state .
feshbach resonances would be of limited importance in cold atom scattering if not for the ability to control the energy of the bare resonance state , and hence the resonance position , through the application of a static magnetic field @xcite . in this context the open and closed channels , in the separated atom limit ,
correspond to pairs of alkali atoms in different combinations of eigenstates of the hyperfine and zeeman hamiltonian with total angular momentum projection quantum numbers @xmath0 and @xmath1 .
the entrance channel is in an @xmath2-wave orbital state ( relative angular momentum quantum number @xmath3 ) , while the closed channels may be in any partial wave , with @xmath4 indicating the projection of the relative orbital angular momentum .
all channels have the same projection of the total angular momentum along the magnetic field axis , @xmath5 , and since their magnetic moments are different , they experience a relative zeeman shift in an applied magnetic field . hence a closed channel bound state
may be shifted in energy relative to the open channel threshold , bringing it into resonance .
the coupling between the entrance channel and @xmath2-wave closed channels arises from the central part of the molecular potential , which only depends on the magnitude of the total electron spin .
therefore , the interatomic potential has short - range off - diagonal terms proportional to the difference between the triplet and singlet potentials .
closed channels of higher angular momentum ( @xmath6 ) can be coupled to the entrance channel through the spin - dipole inteaction @xcite .
the feshbach coupling gives rise to a dispersive variation with magnetic field of the scattering length , well described by @xcite @xmath7.\ ] ] this is associated with a bound molecular state of the coupled channels problem crossing threshold at @xmath8 .
the width of the resonance is characterized by @xmath9 , and @xmath10 is the limiting value of the scattering length for magnetic fields far removed from @xmath11 . in ultracold atomic gases
the @xmath2-wave scattering length gives a complete characterization of the interactions , and ( [ a_vs_b ] ) therefore implies an unmitigated control of the sign and strength of the effective interparticle potential .
this has facilitated the experimental investigation of the crossover between loosely bound cooper pairs and a bose - einstein condensate of true dimers in a trapped fermi gas @xcite .
the coupled channels description of atomic scattering may be solved in full generality once the adiabatic born - oppenheimer ( bo ) curves for the singlet and triplet potentials are known , and gives accurate predictions for the location of the experimentally observed feshbach resonances .
however , there is decided usefulness in the availability of simplified model descriptions , which can be included effortlessly in many - body @xcite and three - body @xcite treatments , or are more amenable to analysis of time - dependent experiments @xcite . fortunately , in most atoms feshbach resonances occur at well separated magnetic fields .
hence , a natural starting point for a simplified description is to consider an isolated resonance . in this paper
, we demonstrate that it is possible to reduce the full multi - channel , low - energy scattering problem of two alkali atoms to an effective two - channel problem for a single feshbach resonance @xcite .
the model consists of a single closed channel containing a bound state , the bare resonance state , which interacts with the scattering continuum in the open channel . in this sense
it is equivalent to fano s theory of resonances arising from a discrete state embedded in the continuum @xcite , but the bare resonance state in the two - channel model is a construct , which may represent the interactions between several closed channels in the full coupled channels problem .
this model is capable of capturing the relevant parts of the full molecular physics of the resonance , provided it is constructed to comply with a small set of low energy parameters .
section [ 2channel_model ] gives an explicit prescription for constructing the two - channel pseudopotentials .
the r - matrix formalism , which is presented in section [ r - matrix_solution ] is an elegant and efficient method of calculating both bound state and scattering properties .
it expresses the logarithmic derivative of the wavefunction at some cutoff radius , beyond which the asymptotic solutions of the schrdinger equation are known , in terms of the solutions of an eigenvalue problem . in section [ numerics ]
we discuss briefly the specific numerical implementation of the method we have used in this paper .
we argue that a discrete variable representation of the hamiltonian combined with a grid based on a finite element partinioning of space defines a numerical procedure , which is efficient in the evaluation of matrix elements and capable of handling the two widely separated length scales associated with the molecular and low energy scattering physics .
section [ results ] gives the results of our calculations for a particular feshbach resonance .
in addition , we give the parameters defining our two - channel model for several other resonances of interest . in all cases
we have ensured that the two - channel model gives an accurate representation of the molecular physics , by comparing with full coupled channels calculations .
we conclude our findings in section [ conclusions ] , where we also contrast our pseudopotential approach with that of other authors .
in the case of an isolated feshbach resonance it is possible to reduce the full coupled channels problem to an effective two - channel model with one open channel ( 1 ) and one closed channel ( 2 ) with a bound state , which interacts with the open channel scattering continuum to produce the resonance .
such a model is capable of capturing the essential physics of the full multi - channel scattering calculation over a wide range of energies . in this section we set up the two - channel model , following the procedure outlined in refs .
@xcite , and discuss how a correspondence may be established between its parametrization and low energy scattering observables , available either through experiments or more exact calculations .
the coupled two - channel hamiltonian describing the relative motion of the particles is chosen to be @xmath12 where @xmath13 is the reduced mass .
we take the zero of energy to correspond to the energy of the dissociated atoms in channel 1 at all magnetic field values .
correspondingly , the only magnetic field dependent component of the problem is the energy offset of the second channel . in the absence of any coupling between the channels the eigenstates of the two - channel problem
is given by the diabatic states @xmath14 , and @xmath15 , where @xmath16 are the eigenstates of the hamiltonian for channel @xmath17 .
the coupling mixes these states , and the resulting dressed ( or adiabatic ) states will in general have components in both channels .
our intention is to solve the coupled channels eigenvalue problem @xmath18 for the components of the wavefunction in each of the two channels , both for the discrete part of the spectrum and for the scattering states . here
@xmath19 is the two by two identity matrix .
the bo potentials are dominated by the van der waals interactions at large internuclear separation , vanishing as @xmath20 , as @xmath21 . at short distances
there is a steep potential barrier .
any potential we choose with these characteristics will give an accurate representation of the molecular physics at low energies .
for simplicity we choose a lennard - jones potential : @xmath22.\ ] ] as in refs .
@xcite we take the potential in both channels to have the same form , but with the closed channel potential shifted upwards in energy , such that its threshold is at @xmath23 , where @xmath24 denotes the difference in magnetic moment between the separated atoms and the bare resonance state .
this simplifies the numerical analysis , and does not seem to have a noticeable effect on the results .
hence , we have @xmath25 finally , we represent the coupling between the two channels by an off - diagonal potential @xmath26 .
again , the specific form of this potential is arbitrary , but it should be a short - range function , since the inter - channel coupling is only significant at small distances .
specifically , we use an exponential form @xmath27 where @xmath28 is the bohr radius .
we thus end up with a five parameter model for the feshbach resonance : @xmath29 and @xmath30 characterize the depth and shape of the diagonal potentials , respectively , while the off - diagonal coupling is of strength @xmath31 .
finally , the offset between the threshold energies of the two channels for a given magnetic field strength is determined by @xmath24 and @xmath32 .
these five parameters are determined by matching the low energy scattering properties of the two - channel model with those of full coupled channels calculations or experimental observables .
the model potentials are illustrated in fig .
[ potentials ] for the case of @xmath33k .
k feshbach resonance at @xmath34 g ( 1 g=0.0001 t ) .
the open potential , @xmath35 is plotted in blue , while the closed channel potential at zero field , @xmath36 , is represented by the red curve .
the diagonal potentials are indistinguishable on this scale .
the interchannel coupling , @xmath26 , is plotted in green .
the inset shows the long - range behavior of the potentials , and the threshold energy @xmath37 is evident as the offset between the asymptotic values of the open and closed channel potentials
. 1 @xmath38 j. ] to make this connection explicit we make use of analytical results for low energy scattering in inverse power law potentials derived by flambaum _
et al_. @xcite . for the open channel , which determines the
far from resonance , or background , scattering properties , we require that our model potential @xmath35 supports @xmath39bound states and that its scattering length equals the background value , @xmath10 , which appears in ( [ a_vs_b ] ) .
we only require that @xmath39 is of the order of the number of bound states in the actual potential , since when @xmath40 , the near - threshold properties are essentially independent of its actual magnitude .
given the atomic mass and the @xmath41 coefficient , which is well known from the asymptotic scattering properties , we may unambiguously construct a model potential meeting these constraints .
both @xmath39 and @xmath10 are related to the wkb phase @xmath42 evaluated at zero energy : @xmath43 , \\
n_s & = & \left [ \frac{\phi}{\pi}-\frac{5}{8 } \right ] + 1,\end{aligned}\ ] ] where @xmath44 $ ] is the integer part , and the mean scattering length ( or characteristic scale length of the van der waals potential ) @xmath45 is governed solely by the asymptotic behavior of the potential . the second equation is necessary to completely determine the semiclassical phase , since the first equation can only be solved for @xmath42 modulo @xmath46 . following flambaum _
et al_. we can then find the relationship between the @xmath47 and @xmath41 coefficients @xmath48 knowing both the @xmath41 and @xmath47 coefficients of our model potential , the range and depth parameters are uniquely determined : @xmath49 the bare resonance state is a particular unit normalized bound state of closed channel potential : @xmath50 \phi_{\rm{res}}(r ) = e_{\rm{res}}(0 ) \phi_{\rm{res}}(r).\ ] ] we arbitrarily choose @xmath51 to be the second to last bound state of the closed channel potential , such that @xmath52 . here
@xmath53 indicates the binding energy of the bare resonance state relative to the threshold in the _ closed _ channel .
this is calculated by diagonalizing @xmath54 .
a calculation of the scattering properties of the two - channel model @xcite reveals that the magnetic field dependent scattering length ( [ a_vs_b ] ) can be related to the strength of the model coupling potential via @xmath55 where @xmath56 .
for @xmath57 the decay width of the quasi - bound molecule giving rise to the resonance is given by @xcite @xmath58 here @xmath59 is the energy normalized outgoing wave scattering state at energy @xmath60 in the presence of the open channel potential alone @xmath61 \phi_{e}(r)=e \phi_{e}(r).\ ] ] its asymptotic form determines the background contribution to the scattering phase shift , @xmath62 @xmath63 the background scattering length follows from @xmath64 in the limit of @xmath65 .
we specify below how the matrix element of the coupling operator in ( [ deltab ] ) is calculated in the r - matrix formalism .
since the scattering length @xmath66 is zero at the magnetic field @xmath67 , the width of the resonance can be either measured or obtained from multi - channel calculations .
hence the strength of the interchannel coupling function may calculated using ( [ deltab ] ) , assuming the difference in magnetic moment of the two channels @xmath24 is known .
we will explain below how this quantity may be obtained .
the measured or calculated position of the feshbach resonance , @xmath11 , is shifted away from the field value @xmath68 , where the bare resonance state crosses zero energy , due to the coupling between the channels .
an accurate analytic estimate of this shift is given by multi - channel quantum defect theory @xcite @xmath69 this provides a link to the threshold energy , @xmath37 . since
the variation of the bare resonance state with magnetic field is given by @xmath70 , it follows from @xmath71 that @xmath72 finally , we address the derivative of the bare resonance state energy with respect to the magnetic field . from the fano theory of resonant scattering @xcite we know that a bound state embedded in the continuum gives rise to a resonant contribution to the phase shift @xmath73 the shift of the resonance position away from the energy of the bare resonance state is @xcite @xmath74 importantly , this shift is independent of the magnetic field .
hence , we expect that for magnetic fields sufficiently far from @xmath11 that the resonance energy is unaffected by threshold effects , the position of the scattering resonance may be fitted with a linear function of the magnetic field , with a slope of @xmath24 .
this defines a procedure for determining the last parameter of our two - channel model . as we shall demonstrate below , this procedure for constructing an effective two - channel model yields predictions , which accurately reproduce the results of a more rigorous calculation for both the continuum _ and _ the bound state spectra
the r - matrix method @xcite has been demonstrated over the past thirty - five years to be a general and quite powerful ab - initio approach to a wide class of atomic and molecular collision problems .
the essential idea is to divide space into two or possibly more physical regions . in each of these regions the time independent schrdinger equation may be solved using techniques designed to be optimal to describe the important physical properties of that region . the solutions and their derivatives are then matched at the region interfaces . a simple example to illustrate this idea is electron scattering from atomic targets . in this case , when the electron is close to the other atomic electrons , there are strong electrostatic , exchange and correlation effects . outside this internal region
, the electron being scattered is subject to purely long - range electrostatic forces .
it then becomes possible to expand the wavefunction in the internal region in a discrete set of eigenstates using the full power of standard bound state approaches .
the internal wavefunction is then matched to known or easily computed solutions of the long - range potentials on the surface of the sphere bounding the two regions .
for the two channel problem considered in this paper , using the definitions in section [ 2channel_model ] , we define the r - matrix eigenstates , \{@xmath75 ( @xmath76 ) } , inside a sphere of radius of @xmath77 , @xmath78 where @xmath79 , the bloch operator @xcite is , @xmath80 the function of the bloch operator is to ensure that any non - zero contributions to matrix elements arising from the kinetic energy operator on the surface of the r - matrix sphere are exactly cancelled by contributions from the bloch operator .
this ensures that the operator @xmath81 is hermitian and that ( [ 2_channel_r_matrix_eqns ] ) has only real eigenvalues .
we use a notational convention where the bound states ( @xmath82 ) have @xmath83 , while the discrete continuum states are labeled with @xmath84 , their number limited by the finite size of the basis .
the r - matrix boundary @xmath85 is chosen large enough that the influence of the potentials at @xmath77 is negligible , such that the r - matrix eigenstates match unto free particle solutions .
these eigenfunctions are then used to expand the solution to the two channel problem at an arbitrary energy , @xmath60 , as @xmath86 the coefficients , @xmath87 , are determined by matching the solution and its first derivative to appropriate asymptotic forms at @xmath77 .
the boundary @xmath85 is chosen sufficiently large that when both channels are closed , corresponding to true vibrational bound states of the coupled system with @xmath82 , the wavefunction has decayed to zero at the boundary .
this ensures that the r - matrix eigenstates reproduce the bound state vibrational spectrum of the molecules precisely . for
the scattering states the matching conditions give , @xmath88 where @xmath89 is the r - matrix for our problem , and @xmath90 is the derivative of the channel wavefunction at the boundary .
it is convenient to define the matrix @xmath91 which determines the energy normalization of the scattering states . here
@xmath92 in the case where channel @xmath93 is open and channel @xmath94 is closed , the asymptotic scattering state for @xmath2-wave collisions can be written as @xmath95 \\ a \exp ( - \kappa r)/r \end{array } \right),\ ] ] where @xmath96 , @xmath97 is an irrelevant normalization constant , and the phase shift is given by @xmath98 . upon use of the r - matrix matching condition
, the phase shift in the open channel is found to be @xmath99 where we have defined an effective open channel r - matrix element , @xmath100 . in the absence of coupling ,
@xmath101 , we have @xmath102 and @xmath103 .
when both channels are open , we must allow for the possibility that the particles may enter the collision region in one channel and exit in the other .
this is described by a @xmath104 k - matrix , @xmath105 , and the asymptotic scattering state becomes in a general matrix notation @xmath106.\ ] ] the off - diagonal elements of @xmath107 describe the cases where the incoming channel and outgoing channel differ .
the matrices @xmath108 and @xmath109 give the regular and irregular solutions in the asymptotic region @xcite . the r - matrix matching at @xmath77 then gives a linear system of equations for the multi - channel k - matrix : @xmath110 { \bf k } = [ { \bm { \mathcal{r } } } { \bf k}^{-1/2 } { \bf f'}(a ) - { \bf k}^{-1/2 } { \bf f}(a ) ] .\ ] ] the s - matrix can be found from @xmath111 the multi - channel s - matrix satisfies the unitarity condition @xmath112 for all @xmath113 . a great advantage of the r - matrix method can be seen from ( [ r - matrix_element ] ) .
the explicit energy dependence enables the r - matrix to be calculated quite efficiently with energy . in the presence of resonances ,
the phase shift varies rapidly with energy but the corresponding r - matrix is constructed easily from energy independent parts . in addition , the r - matrix poles are often very good approximations to the positions of the resonances .
the widths may be extracted in a variety of ways , the simplest being computing the phase shift as a function of energy and fitting the result to a breit - wigner form plus background term .
this makes r - matrix theory ideal for the study of magnetic field induced feshbach resonances . the original derivation of the r - matrix method by wigner and eisenbud @xcite used r - matrix states satisfying a fixed , zero derivative boundary condition at the r - matrix radius .
while this is perfectly acceptable formally , it is a very slowly convergent process numerically as a consequence of the fact that the scattering wavefunction at energy @xmath60 satisfies a different boundary condition at the r - matrix radius .
thus , the use of fixed boundary condition states , can only be correct in the limit of a complete expansion basis .
the most appropriate way to look at the problem from a variational viewpoint is that we are finding the rayleigh - ritz solution to the self - adjoint operator @xmath114 . by chosing a basis which does not satisfy a fixed derivative condition at the surface , convergence of the solution
is quickly achieved with an incomplete basis .
it is possible to devise an approximate correction procedure @xcite to account for the infinity of neglected levels using a fixed boundary condition basis . in the approach above
, the use of the bloch operator eliminates the need for specifying a specific derivative condition on the basis .
in fact , basis functions with arbitrary derivatives may be used as an expansion set .
this leads to a very rapid convergence of the s - matrix as a function of the basis set size .
the higher lying levels have little physical significance individually , but their spectral sum is effectively converging to the r - matrix .
in addition , it can be shown that truncating the very high lying r - matrix levels has a negligible effect on the low - energy scattering properties .
this is in contrast to some other approaches @xcite where the elimination of a single state gives completely erroneous results .
in the r - matrix method as outlined in the last section , the critical element is the computation of the r - matrix eigenvalues and eigenvectors .
the bloch operator formalism enables the use of powerful variational methods , which do not require the underlying basis set to satisfy any specific boundary condition .
this produces a very rapid convergence of the results with respect to basis set size .
the choice of primitive basis in which to expand the r - matrix eigenstates is , of course , arbitrary . in practice
many choices are eliminated on the basis of computational convenience or efficiency . for the problem under consideration in this paper
, there is the additional constraint that the basis accurately describe both the two - body vibrational bound states as well as the very low energy continuum states .
for the calculations presented here this requires a highly accurate description of the wavefunction to internuclear distances of approximately 1000@xmath115 .
the finite element discrete variable ( fedvr ) basis , described in other publications @xcite , has proven ideal for this purpose .
while a detailed description of the method must be relegated to the references , a few qualitative remarks here are useful .
the essential features of the fedvr method are 1 ) to subdivide space into a variable number of elements of different length and 2 ) to use a flexible number of dvr basis functions in each element to accommodate the local spatial variations of the wavefunction .
the basis functions in each element are chosen to be the lagrange interpolation polynomials at the gauss - legendre - lobatto quadrature points in that element .
the lobatto quadrature rule , which forces a common quadrature point at each boundary , enables the construction of a `` bridge '' basis function which ensures wavefunction continuity ( communication ) across the elements .
the `` bridge '' function is simply a linear combination of the last dvr function in region @xmath17 and the first dvr function in region @xmath116 .
note , that the order of the dvr basis in each element need not be the same . in addition , the use of a dvr basis in each element , means that matrix elements of the potential are simply the value of the potential at the gauss - legendre - lobatto quadrature points in that element .
the components of the kinetic energy matrix in each element are given by analytic expressions involving the derivatives of the lagrange interpolating polynomials evaluated at the quadrature points .
first and higher order derivatives of the basis set are not and need not be continuous across the elements for the numerical procedure to be convergent in the limit of a complete basis .
the procedure enables us to capture both the short and long range features of the molecular wavefunctions with a reasonably small basis ( @xmath117300 functions / channel ) .
we thus have a large number of points in the potential region , where the wavefunction oscillates rapidly , and an increasingly sparse grid , as one gets deeper into the asymptotic region where the behavior of the wavefunction is governed purely by wavelength ( energy ) considerations .
finally , an appealing attribute of the fedvr basis for r - matrix calculations is that only a single fedvr function contributes to the r - matrix elements at the boundary .
this is a consequence of the fact that the dvr functions are either zero or one at the fedvr points .
in this section we present results of calculations within our two - channel model , which fully illustrate the resonance scattering and molecular physics of the feshbach resonance . in table
[ parameters ] the parameters defining the minimal two - channel model description and the set of physical parameters on which it is based are given for several feshbach resonances of experimental interest .
we present a comprehensive comparison of our two - channel results with those of a full coupled channels calculation only for the specific example of the feshbach resonance occuring at @xmath11=202.15 g in fermionic @xmath33k in the @xmath118 and @xmath119 hyperfine states .
however , we have checked that our model is in agreement with more elaborate calculations for all the cases quoted in the table .
species & @xmath41 ( au ) & @xmath10 ( au ) & @xmath39 & @xmath11 ( g ) & @xmath9 ( g ) & @xmath24 ( mhz / g ) & @xmath29 ( au ) & @xmath30 ( au ) & @xmath31 ( au ) & @xmath37 ( au ) + @xmath33k & 3897 & 172.06 & 27 & 202.15 & 7.788 & 2.35 & 1.631@xmath120 & 9.177 & 0.221 & 1.192@xmath121 + @xmath33k & 3897 & 172.15 & 27 & 224.21 & 8.12 & 2.35 & 1.631@xmath120 & 9.177 & 0.226 & 1.112@xmath121 + @xmath122rb & 4700 & -480 & 35 & 155.199 & 10.641 & -3.345 & 1.124@xmath120 & 10.075 & 0.203 & 3.283@xmath121 + the essential molecular physics of the feshbach resonance is illustrated in fig .
[ e_multiple ] , where the energies of the last two bound states of the coupled channels problem are plotted ( solid lines ) .
the energies of the uncoupled diabatic states corresponding to the bare feshbach molecule with slope @xmath24 and the last bound state of the open channel at energy @xmath123 are indicated by the dashed lines . in this case
the coupling shifts the feshbach resonance position downwards from the zero crossing of the bare feshbash state at @xmath124=211.39 g to @xmath11 , where the binding energy of the last bound state at @xmath125=0 vanishes , and this state enters the continuum .
the last bound state of the open channel is seen to play a pivotal role , as it interacts strongly with the bare resonance state , leading to an avoided crossing .
hence the dressed molecules produced in an adiabatic sweep of the magnetic field across the feshbach resonance , starting at @xmath126 , will evolve into the last bound state of the open channel potential ( magenta curve ) for @xmath127 , not the bare resonance state in the closed channel .
instead it is the dressed state with energy @xmath128 ( green curve ) which tracks @xmath129 in this regime . at fields
much larger than @xmath11 the energy of this state approaches @xmath123 .
this agrees with earlier observations by other authors @xcite . for other resonances , such as
the two in @xmath130li , the last bound state of the open channel lies much deeper , and has no effect on the feshbach molecules . the open circles in fig . [ e_multiple ]
give the results of a full coupled channels calculation .
we observe that our simple two - channel model gives a faithful representation of the dressed states .
the figure also shows the discrete r - matrix representation of the scattering continuum . at low energies they bear some semblance to the eigenstates in a spherical box with zero derivative on the boundary , though no such correspondence can be established at higher energies .
their spectral density is therefore tied to the size of the box . in particular , the first continuum state is not at exactly zero energy , but is displaced a small positive amount , representing the zero point energy in the effective box potential .
the scattering resonance is visible as a scar across the continuum part of the spectrum .
we remark that while the energy of the second to last adiabatic bound state approaches that of the bare resonance state far below the resonance , it is in general affected by the coupling to deeper bound states of the open channel potential and in fact never becomes linear in the magnetic field .
it is therefore important to obtain the magnetic moment difference by fitting the position of the scattering resonance for fields above @xmath11 .
k feshbach resonance at @xmath34 g. the @xmath131 ( magenta line ) and @xmath128 ( green line ) bound states of the two - channel model as counted at @xmath132 , are compared with the results of a full multi - channel calculation ( open symbols ) . at the resonance position @xmath8 the number of bound states changes by one .
the low energy r - matrix eigenstates with @xmath57 ( blue lines ) correspond to the eigenstates of a spherical box with vanishing derivative at the boundary . at higher energies
no such correspondence exists ( see text ) .
the bare resonance state and the last bound state of the open channel potential with energy @xmath133 mhz are indicated by the dashed lines . ]
figure [ z_multiple ] gives additional details of the behavior of the dressed states as the magnetic field is varied .
the figure displays the weight of the dressed state wavefunctions in the closed channel , defined as @xmath134 in accordance with the bound state spectrum in fig .
[ e_multiple ] the last adiabatic bound state is a pure open channel state in both the high and low field limit .
near the feshbach resonance it acquires some closed channel character , but it remains dominated by the open channel . in contrast , the bound state with energy @xmath128 changes from the open channel state with energy @xmath123 to the bare resonance state in the closed channel , as the magnetic field is changed from above to below @xmath11 @xcite . again
, the results of a coupled channels calculation are shown for comparison , and are found to confirm the simple model . for the coupled channels calculation @xmath135 represents the sum of contributions from all the closed channels coupled with the open channel .
the variation of @xmath135 with @xmath125 for the lowest lying states in the continuum shows that as the resonance energy increases with increasing magnetic field , the discrete feshbach state becomes less diluted by the coupling to the continuum . .
note that for the molecular state created by a downward sweep across the feshbach resonance the population in the closed channel never exceeds @xmath136 .
the closed channel population for the last two bound states resulting from a cc calculation are shown for comparison .
continuum curves with peak positions at higher magnetic fields correspond to r - matrix eigenstates with a higher energy . ] to extract information about the low energy scattering in the two - channel model , we find the scattering phase shift as a function of energy using ( [ tan_delta ] ) .
fitting to the effective range expansion @xmath137 then gives the magnetic field dependent scattering length and effective range .
these are plotted in fig .
[ scattering ] , and the scattering length is seen to be in agreement with ( [ a_vs_b ] ) , as expected since the two - channel model has been constructed to reproduce the salient features of this expression . for reference
the effective range is also plotted . for most of the magnetic field interval
shown @xmath138 remains small and positive , but it diverges at @xmath67=209.93 g , where the scattering length vanishes @xcite . ) , shown by the solid line .
] the energy width of the quasi - bound dressed feshbach state as determined by ( [ gamma ] ) is plotted in fig .
[ ewidth ] . as the energy approaches zero @xmath139 as required by the wigner threshold law . using ( [ deltab ] )
it follows that @xmath140 this wigner threshold regime limit is also indicated in fig .
[ ewidth ] .
k as a function of the scattering energy on a , logarithmic scale . the two - channel expression ( [ gamma ] ) ( solid line ) is compared to its limit in the wigner threshold regime ( [ gamma_lowe ] ) ( dashed line ) . ] to illustrate in greater detail how the scattering evolves with energy and magnetic field we plot in fig .
[ t_matrix ] the energy dependent @xmath141-matrix , @xmath142 , for several different values of the magnetic field .
we only plot the purely open channel component of @xmath143 , which is related to the phase shift via @xmath144 the total phase shift may be written as a sum of a background and the resonant contribution @xcite @xmath145 for the resonance under consideration here we may take the background term to be independent of the magnetic field , but this is not always the case ( see _ e.g. _
@xcite ) . for each magnetic field
the figure compares the energy variation of both the real and imaginary parts of @xmath146 calculated within the two - channel model of this paper and full coupled channels calculations .
for completeness we also plot for both models the resulting @xmath147 . at low values of @xmath125
the feshbach resonance has almost no influence on the scattering , which is dominated by the background contribution .
right on the feshbach resonance the scattering near threshold is strongly affected by the coupling to the closed channel bound state . in the last four panels
the energy of the bare resonance state is indicated with a vertical dashed black line .
when it lies in the continuum it gives rise to a scattering resonance , the shape of which depends value of the background phase shift in the neighborhood of the resonance position . at @xmath148
g the resonance is broad ( corresponding to a finite value of @xmath149 ) , and its position does not line up with the energy of the bare resonance state . at higher fields the resonance follows the bare resonance state .
the last three panels demonstrate the interplay between the two different contributions to the scattering by plotting @xmath146 at magnetic field values , where the resonance occurs at energies such that the background phase shift modulo @xmath46 is close to @xmath150 , @xmath151 , and @xmath152 , respectively . as the scattering energy
is increased through the resonance the phase shift increases by approximately @xmath46 .
correspondingly , the values of the real and imaginary parts of @xmath146 reflect the value of the background phase shift on resonance .
this gives @xmath153 , @xmath154 , and @xmath155 for the real part of the @xmath141-matrix on resonance , while the imaginary part at the same magnetic fields is @xmath156 , @xmath157 , and @xmath158 , in good agreement with the numerical findings .
we have presented a recipe for constructing a simplified two - channel model of an isolated feshbach resonance . given a set of five physical parameters , which can be obtained from either experiments or exact calculations ,
the parameters characterizing our model can be derived without ambiguity . within the r - matrix formalism , the bound state part of the spectrum as well as the states scattering at arbitrary energies in the continuum are obtained from a diagonalization of the two - channel hamiltonian ( including the bloch operator ) .
the fedvr is particularly well suited to represent the model hamiltonian , since the combination of short range molecular potentials with long wave length scattering requires a spatial grid which encompasses two widely separated length scales .
we have provided the parameters defining the two - channel model for several feshbach resonances used in current experiments , and made direct comparison between the predictions of our model with the results of exact calculations .
this comparison shows that over a large range of energies and magnetic fields the two - channel model reliably reproduces the results of full coupled channels calculations .
the usefulness of the two - channel model lies in its simplicity .
in particular , the magnetic field enters only as a parameter in the closed channel potential .
hence the stationary two - channel hamiltonian provides an excellent starting point for studies of propagation in the presence of a time dependent magnetic field .
such time - dependent problems are also well suited for the fedvr method @xcite .
we end by contrasting the two - channel scheme of this paper with simplified models of feshbach resonances developed by other authors
. there are examples of a number of similar two - channel models in the literature @xcite .
the new elements of this work are the use of pseudopotentials , which gives as realistic a representation of the molecular potentials of the full coupled channels problem as possible , and a detailed prescription for generating them .
specifically , we choose model potentials which support the same number of bound states as the entrance channel in the full problem .
we do not anticipate that the additional details of the model presented here will lead to fundamentally different results .
but it does lead to a more accurate representation of the dressed state wavefunctions .
similar remarks apply when our model is compared with those in which even simpler pseudopotentials are introduced , such as a well with a barrier @xcite , two coupled wells @xcite or zero range potentials @xcite .
these approaches have the advantage that they are amenable to analytical treatment , but they give no details of the molecular physics at short range .
the virtue of a model description is measured on its ability to reproduce the available data effectively , while illuminating the important underlying physical principles .
we believe that our approach coupled with an r - matrix analysis implemented within a fedvr scheme constitutes an efficient and transparent procedure for generating the bound state and scattering properties of the coupled channels feshbach problem .
h. feshbach , ann .
( n.y . ) * 5 * , 357 ( 1958 ) ; * 19 * , 287 ( 1962 ) .
e. tiesinga , b. j. verhaar and h. t. c. stoof , phys . rev . a * 47
* , 4114 ( 1993 ) .
a. j. moerdijk , b. j. verhaar and a. axelsson , phys .
a * 51 * , 4852 ( 1995 ) .
h. t. c. stoof , j. m. v. a. koelman and b. j. verhaar , phys .
rev b * 38 * , 4688 ( 1988 ) .
k. m. ohara , s. l. hemmer , m. e. gehm , s. r. granade and j. e. thomas , science * 298 * , 2179 ( 2002 ) . c. a. regal , m. greiner and d. s. jin , phys .
lett . * 92 * , 040403 ( 2004 ) .
m. w. zwierlein , c. a. stan , c. h. schunck , s. m. f. raupach , a. j. kerman and w. ketterle , phys .
92 * , 120403 ( 2004 ) .
m. bartenstein , a. altmeyer , s. riedl , s. jochim , c. chin , j. hecker denschlag and r. grimm , phys .
* 92 * , 120401 ( 2004 ) . j. kinast , s. l. hemmer , m. e. gehm , a. turlapov , and j. e. thomas , phys .
* 92 * , 150402 ( 2004 ) .
t. bourdel , l. khaykovich , j. cubizolles , j. zhang , f. chevy , m. teichmann , l. tarruell , s. j. j. m. f. kokkelmans and c. salomon , phys .
lett . * 93 * , 050401 ( 2004 ) .
k. gral , t. khler , s. gardiner , e. tiesinga and p. julienne , j. phys .
b * 37 * , 3457 ( 2004 ) .
t. khler , k. gral and t. gasenzer , phys .
a * 70 * , 023613 ( 2004 ) .
s. simonucci , p. pieri and g. c. strinati , europhys . lett . * 69 * , 713 ( 2005 )
. m. h. szymaska , k. gral , t. khler and k. burnett , phys .
a * 72 * , 013610 ( 2005 ) .
m. stoll and t. khler , phys .
a * 72 * , 022714 ( 2005 ) .
f. h. mies , e. tiesinga and p. s. julienne , phys . rev .
a * 61 * , 022721 ( 2000 ) .
t. khler , t. gasenzer , p. s. julienne and k. burnett , phys .
* 91 * , 230401 ( 2003 ) .
u. fano , phys .
rev . * 124 * , 1866 ( 1961 ) .
v. v. flambaum , g. f. gribakin and c. harabati , phys .
a * 59 * , 1998 ( 1999 ) .
a comprehensive and up to date article , to be published in advances in atomic and molecular physics is , p. g. burke , c. j. noble and v. m. burke , `` r - matrix theory of atomic , molecular and optical processses '' b. i. schneider , `` the r - matrix theory of electron - molecule scattering '' , in , electron - atom and electron - molecule collisions , ed .
by j. hinze , plenum press , new york and london ( 1983 ) p. s. julienne , e. tiesinga and t. khler , j. mod .
optics * 51 * , 1787 ( 2004 ) . c. bloch , nucl .
phys . * 4 * , 503 ( 1957 ) .
n. f. mott and h. s. w. massey , _ the theory of atomic collisions _
( oxford university press , oxford , 1965 ) .
e. wigner and l. eisenbud , phys .
* 72 * , 29 ( 1947 ) p. j. a. buttle , phys
. rev . * 160 * , 719 ( 1967 ) o. i. tolstikhin , v. n. ostrovsky and h. nakamura , phys .
a * 58 * , 2077 ( 1998 ) .
t. n. rescigno and c. w. mccurdy , phys .
a * 62 * , 032706 - 1 ( 2000 ) b. i. schneider and l. collins , j. non - cryst .
solids * 351 * , 1551 ( 2005 ) .
see epaps document no .
[ _ number will be inserted by publisher _ ] for animations showing the variation of the open and closed channel components of the @xmath159 and @xmath160 dressed state wavefunctions with magnetic field , as well as the magnetic field dependence of the @xmath141-matrix .
this document can be reached via a direct link in the online article s html reference section or via the epaps homepage ( http://www.aip.org/pubservs/epaps.html ) .
b. gao , phys .
a * 58 * , 4222 ( 1998 ) .
m. bartenstein , a. altmeyer , s. riedl , r. geursen , s. jochim , c. chin , j. hecker denschlag , r. grimm , a. simoni , e. tiesinga , c. j. williams and p.
s. julienne , phys .
lett . * 94 * , 103201 ( 2005 ) .
s. de palo , m. l. chiofalo , m. j. holland and s. j. j. m. f. kokkelmans , phys .
a * 327 * , 490 ( 2004 ) .
s. j. j. m. f. kokkelmans , j. n. milstein , m. l. chiofalo , r. walser and m. j. holland , phys .
a * 65 * , 053617 ( 2002 ) . c. chin , cond - mat/0506313 ( 2005 ) .
m. g. moore , cond - mat/0506383 ( 2005 ) .
@xmath161{fig6a.eps } & \includegraphics[scale=0.4]{fig6b.eps } \\ \includegraphics[scale=0.4]{fig6c.eps } & \includegraphics[scale=0.4]{fig6d.eps } \\ \includegraphics[scale=0.4]{fig6e.eps } & \includegraphics[scale=0.4]{fig6f.eps } \\ \end{array } $ ] | a feshbach resonance arises in cold atom scattering due to the complex interplay between several coupled channels .
however , the essential physics of the resonance may be encapsulated in a simplified model consisting of just two coupled channels . in this paper
we describe in detail how such an effective feshbach model can be constructed from knowledge of a few key parameters , characterizing the atomic born - oppenheimer potentials and the low energy scattering near the resonance .
these parameters may be obtained either from experiment or full coupled channels calculations . using r - matrix theory
we analyze the bound state spectrum and the scattering properties of the two - channel model , and find it to be in good agreement with exact calculations . |
extreme horizontal - branch ( ehb ) stars occupy the hot end of the horizontal branch ( hb ) in globular clusters with extended blue hb tails .
the envelope masses of these stars are too small ( @xmath2 @xmath30.02 @xmath4 ) to sustain hydrogen - shell burning , and thus nearly all of their luminosity comes from helium burning in the core .
recent observations have discovered an unexpected population of hot stars lying below the canonical ehb in the uv color - magnitude diagram ( cmd ) of @xmath0 cen ( dcruz et al .
2000 and references therein ) . in the present paper
we will use new uv observations of the globular cluster ngc 2808 to explore the origin of these subluminous ehb stars and will suggest that these stars may be the progeny of stars which underwent extensive mixing during a delayed helium flash on the white - dwarf ( wd ) cooling curve ( brown et al .
2001 ) .
our data for ngc 2808 were obtained in the far - uv ( fuv , @xmath5 ) and near - uv ( nuv , @xmath6 ) bandpasses of the space telescope imaging spectrograph ( stis ) .
the hb of ngc 2808 is bimodal with a large gap between the blue hb ( bhb ) and red hb ( rhb ) stars .
in addition , ngc 2808 has a very long blue hb tail that is punctuated by two gaps : one between the ehb and bhb and one within the ehb itself ( sosin et al .
1997 ; walker 1999 ; bedin et al .
our stis cmd ( figure 1 ) shows the following features : * the gap between the ehb and bhb at @xmath7 mag is well detected , as is the gap between the bhb and rhb at @xmath8 mag .
the gap within the ehb seen in optical cmds is not present .
* there is a large population of hot subluminous hb stars , previously known to exist only in @xmath0 cen ( dcruz et al .
2000 ) . out of a total of 75 ehb stars ,
46 are fainter than the canonical zero - age horizontal branch ( zahb ) . *
5 luminous post - ehb stars are found at @xmath9 mag .
to study the origin of the hot subluminous ehb stars , we computed a set of evolutionary sequences which followed the evolution of a low - mass star continuously from the main sequence through the helium flash to the zahb .
all sequences had the same heavy - element abundance z of 0.0015 , corresponding to [ fe / h ] = @xmath101.31 for [ @xmath11/fe ] = 0.3 , and an initial main - sequence mass @xmath12 of @xmath13 , corresponding to an age of 13 gyr at the tip of the rgb .
the sequences only differed in the extent of mass loss along the rgb which we specified by varying the mass - loss parameter @xmath14 in the reimers formulation from 0 to 1 .
our purpose was to determine if mixing between the helium core and hydrogen envelope might occur during the helium flash , as found previously by sweigart ( 1997 ) .
a representative sample of our evolutionary tracks is shown in figure 2 .
for @xmath15 the helium flash occurs at the tip of the rgb . as the mass loss increases , however , the models peel off the rgb and evolve to high effective temperatures before igniting helium ( so - called `` hot he - flashers '' ; castellani & castellani 1993 ; dcruz et al . 1996 ) .
for @xmath16 the helium flash occurs between the tip of the rgb and the top of the wd cooling curve . in all sequences with @xmath17 the flash convection zone produced by the high helium - burning luminosity ( peak @xmath18 ) failed to reach the hydrogen envelope by @xmath31 pressure - scale height . thus mixing between the helium core and the hydrogen envelope
does not occur if a star ignites helium either on the rgb or during the evolution to the top of the wd cooling curve . in these cases
we confirm the canonical assumption that the helium flash does not affect either the envelope mass or composition .
the canonical evolution described above changes dramatically when the helium flash occurs further down the wd cooling curve ( figure 2 ) . as a star descends the cooling curve , the entropy barrier of its hydrogen shell decreases ( iben 1976 ) . as a result
, the flash convection is then able to penetrate deeply into the hydrogen envelope ( sweigart 1997 ; brown et al . 2001 ) .
the protons captured by the flash convection will be mixed into the helium - burning core while helium and carbon from the core will be mixed outward into the envelope .
the calculations of sweigart ( 1997 ) indicate that this `` flash mixing '' will consume most , if not all , of the envelope hydrogen while simultaneously enriching the envelope in helium and carbon .
all of our sequences with @xmath19 encountered flash mixing .
these sequences were stopped at the onset of mixing due to the numerical difficulty of following the proton mixing and nucleosynthesis .
sequences with @xmath20 did not ignite helium and thus died as helium white dwarfs .
flash mixing is a consequence of the basic properties of the stellar models and hence should occur whenever a star ignites helium on the wd cooling curve .
analogous mixing occurs during a very late helium - shell flash according to the `` born - again '' scenario for producing hydrogen - deficient stars ( iben 1995 and references therein ) .
ehb evolutionary tracks for both canonical ( @xmath17 ) and flash - mixed ( @xmath19 ) sequences are plotted in figure 3 . the canonical ( i.e. , unmixed ) models have the same h - rich envelope composition as the pre - helium flash models .
the zahb models for the flash - mixed tracks were obtained by changing the envelope composition to mimic the effects of flash mixing .
we considered both he + c - rich and he - rich envelope compositions for these flash - mixed models ( see caption of figure 3 ) . for comparison
the bottom panel of figure 3 shows a set of tracks with @xmath19 which have the same h - rich envelope composition as the canonical models .
several features of figure 3 deserve comment . first , there is a well - defined high temperature limit to the canonical ehb at an effective temperature @xmath21 of @xmath331,500 k for the present metallicity . within the canonical framework
it is not possible to produce hotter ehb stars regardless of the extent of mass loss along the rgb .
second , the flash - mixed tracks are separated by a gap of @xmath36000 k in @xmath21 from the hot end of the canonical ehb . as a consequence , flash mixing leads to a sharp dichotomy in the properties of the ehb stars .
the change from canonical to flash - mixed evolution occurs over an interval of only 0.001 in @xmath14 , corresponding to a mass - loss difference of only 10@xmath22 .
in order to compare our ehb models with the stis data , we computed new stellar atmospheres for both h - rich and h - depleted compositions . the spectral energy distribution for a canonical ehb star is compared to the spectra for 2 flash - mixed stars with he + c - rich and he - rich atmospheres in figure 4 .
the enhanced helium abundance in the flash - mixed stars reduces the hydrogen opacity below 912 so that more of the flux is radiated in the extreme uv at the expense of the flux at longer wavelengths .
enhancing the carbon abundance along with helium restores some of this extreme uv opacity , but the resulting spectrum is still redder and fainter in the far - uv than a normal spectrum . using these he + c - rich and he - rich stellar atmospheres , we translated the zahb models from the flash - mixed sequences to the stis cmd ( see figure 5 ) .
the flash - mixed models show the same dramatic drop in the far - uv luminosity as the stis data .
this drop can be attributed to 1 ) the lower far - uv flux in the spectral energy distribution of the flash - mixed models ( see figure 4 ) and 2 ) the larger bolometric correction due to their higher temperatures ( see figure 3 ) .
evolution off the zahb will fill in the region between these flash - mixed models in figure 5 and the canonical zahb .
radiative levitation of fe would shift the flash - mixed models redward , possibly explaining the color spread of the subluminous stars .
in contrast , the @xmath14 = 0.860 zahb model with a normal h - rich envelope in figure 5 lies just @xmath30.1 mag fainter than the hot end of the canonical ehb , a consequence of its slightly smaller core mass ( @xmath23 ) .
it is clear therefore that hot he flashers which ignore flash mixing can not explain the subluminous stars in either @xmath0 cen or ngc 2808 .
optical cmds also show a gap within the ehb of ngc 2808 .
we have plotted our zahb models in the ( @xmath24 , @xmath25 ) cmd to see if the temperature difference between the flash - mixed and canonical tracks in figure 3 might produce such a gap . as expected ,
the flash - mixed models are separated by a gap of @xmath30.5 mag in v from the hot end of the canonical ehb .
our simulations , which include evolution off the zahb , predict an ehb gap close to that observed in ngc 2808 ( see brown et al .
a similar ehb gap is not seen in @xmath0 cen , possibly because it has been obscured by the metallicity spread of the cluster .
* stars which ignite helium on the white - dwarf cooling curve will undergo substantial mixing between the helium core and hydrogen envelope during the helium flash .
this flash - induced mixing will greatly enhance the envelope helium and carbon abundances .
* flash - mixed ehb stars will appear subluminous in uv color - magnitude diagrams due to their redistributed far - uv flux and larger bolometric corrections , in agreement with the subluminous stars in @xmath0 cen and ngc 2808 .
* flash mixing leads to a dichotomy in the properties of ehb stars .
the temperature gap between the flash - mixed and canonical ( i.e. , unmixed ) stars may explain the ehb gap found in optical cmds of ngc 2808 .
* flash mixing on the white - dwarf cooling curve may provide a new evolutionary channel for producing h - deficient stars , particularly the high gravity , he - rich sdo stars ( lemke et al . 1997 ) and the minority of sdb stars that are he - rich ( moehler , heber , & durrell 1997 ) .
bedin , l. r. , piotto , g. , zoccali , m. , stetson , p. b. , saviane , i. , cassisi , s. , & bono , g. 2000 , , 363 , 159 brown , t. m. , sweigart , a. v. , lanz , t. , landsman , w. b. , & hubeny , i. 2001 , , in press ( astro - ph/0108040 ) castellani , m. , & castellani , v. 1993 , , 407 , 649 dcruz , n. l. , dorman , b. , rood , r. , & oconnell , r. 1996 , , 466 , 359 dcruz , n. l. , oconnell , r. w. , rood , r. t. , whitney , j. h. , dorman , b. , landsman , w. b. , hill , r. s. , stecher , t. p. , & bohlin , r. c. 2000 , , 530 , 352 iben , i. , jr .
1976 , , 208 , 165 iben , i. , jr .
1995 , phys .
, 250 , 1 lemke , m. , heber , u. , napiwotzki , r. , dreizler , s. , & engels , d. 1997 , in the third conference on faint blue stars , ed .
a. g. d. philip , j. w. liebert & r. a. saffer ( schenectady : l. davis press ) , 375 moehler , s. , heber , u. , & durrell , p. r. 1997 , , 317 , l83 sosin , c. , dorman , b. , djorgovski , s. g. , piotto , g. , rich , r. m. , king , i. r. , liebert , j. , phinney , e. s. , & renzini , a. 1997 , , 480 , l35 sweigart , a. v. 1997 , in the third conference on faint blue stars , ed .
a. g. d. philip , j. w. liebert & r. a. saffer ( schenectady : l. davis press ) , 3 walker , a. r. 1999 , , 118 , 432 | hot subluminous stars lying up to 0.7 mag below the extreme horizontal branch ( ehb ) are found in the ultraviolet ( uv ) color - magnitude diagrams of both @xmath0 cen ( dcruz et al .
2000 ) and ngc 2808 ( brown et al . 2001 ) . in order to explore the evolutionary status of these subluminous stars ,
we have evolved a set of low - mass stars continuously from the main sequence through the helium - core flash to the hb for a wide range in the mass loss along the red - giant branch ( rgb ) .
stars with the largest mass loss evolve off the rgb to high effective temperatures before igniting helium in their cores .
our results indicate that the subluminous ehb stars , as well as the gap within the ehb of ngc 2808 , can be explained if these stars undergo a late helium - core flash while descending the white - dwarf cooling curve . under these conditions
the convection zone produced by the helium flash will penetrate into the stellar envelope , thereby mixing most , if not all , of the envelope hydrogen into the hot helium - burning interior , where it is rapidly consumed ( sweigart 1997 ) .
this phenomenon is analogous to the `` born - again '' scenario for producing hydrogen - deficient stars following a very late helium - shell flash .
this `` flash mixing '' of the stellar envelope greatly enhances the envelope helium and carbon abundances and , as a result , leads to a discontinuous jump in the hb effective temperature .
we argue that the ehb gap in ngc 2808 is associated with this theoretically predicted dichotomy in the hb morphology . using new helium- and carbon - rich stellar atmospheres , we show that these changes in the envelope abundances of the flash - mixed stars will suppress the uv flux by the amount needed to explain the hot subluminous ehb stars in @xmath0 cen and ngc 2808 .
moreover , we demonstrate that models without flash mixing lie , at most , only @xmath1 mag below the ehb , and hence fail to explain the observations .
flash mixing may also provide a new evolutionary channel for producing the high gravity , helium - rich sdo and sdb stars .
# 1_#1 _ # 1_#1 _ = # 1 1.25 in .125 in .25 in |
recent years have seen a surge of modelling particle dynamics from a discrete potential theoretic perspective .
the models , which usually run under the heading _ aggregation models _ , in particular cases boil down to ( harmonic / poisson ) redistribution of a given initial mass ( sandpile ) according to some prescribed governing rules .
the most famous and well - known model is the poincar s balayage , where a given initial mass distribution @xmath2 ( in a given domain @xmath3 ) is to be redistributed ( or mapped ) to another distribution @xmath4 on the boundary of the domain @xmath5 where @xmath4 is uniquely defined through @xmath2 , and @xmath3 .
this model uses continuous amounts of mass instead of discrete ones ; the latter is more common in chip firing on graphs ( see @xcite for instance ) . a completely different model called _ partial - balayage _ ( see @xcite ) aims at finding a body ( domain ) that is gravi - equivalent with the given initial mass .
this problem , in turn , is equivalent to variational inequalities and the so - called obstacle problem .
the discrete version of this problem was ( probably for the first time ) studied by d. zidarov , where he performed ( what is now called ) a _ divisible sandpile _ model ; see @xcite page 108 - 118 .
levine @xcite , and levine - peres @xcite , @xcite started a systematic study of such problems , proving , among other things , existence of scaling limit for divisible sandpiles .
although zidarov was the first to consider such a problem , the mathematical rigour is to be attributed to levine @xcite , and levine - peres @xcite , @xcite .
the divisible sandpile , which is of particular relevance to our paper , is a growth model on @xmath0 @xmath6 which amounts to redistribution of a given continuous mass .
the redistribution of mass takes place according to a ( given ) simple rule : each lattice point can , and eventually must topple if it already has more than a prescribed amount of mass ( sand ) .
the amount to topple is the excess , which is divided between all the neighbouring lattice points equally or according to a governing background pde . the scaling limit of this model , when the lattice spacing tends to 0 , and the amount of mass is scaled properly , leads to the _ obstacle problem _ in @xmath7 @xmath6 .
the divisible sandpile model of zidarov , and levine - peres also relates to a well - known problem in potential theory , the so - called quadrature domains ( qd ) @xcite . a quadrature domain @xmath3 ( with respect to a given measure @xmath2 ) is a domain that has the same exterior newtonian potential ( with uniform density ) as that of the measure @xmath2 .
hence , potential theoretically @xmath2 and @xmath8 are equivalent in the free space ; i.e. outside the support of @xmath3 one has @xmath9 , where these are the newtonian potentials of @xmath2 , respectively @xmath8 .
the odometer function @xmath10 of levine - peres ( which represents the amount of mass emitted from each lattice site ) corresponds to the difference between the above potentials ( up to a normalization constant ) , i.e. @xmath11 , where @xmath10 is @xmath12 and @xmath13 in @xmath14 .
this , expressed differently , means that @xmath15 for all @xmath16 harmonic and integrable over @xmath3 ( see @xcite ) .
in many other ( related ) free boundary value problems ( known as bernoulli type ) the zero boundary gradient @xmath17 in the above takes a different turn , and is a prescribed ( strictly ) non - zero function , and the volume potential @xmath18 in is replaced by surface / single layer potential ( of the a priori unknown domain ) @xmath19 . in terms of sandpile redistribution
this means to find a domain @xmath3 such that the given initial mass @xmath2 in is replaced by a prescribed mass @xmath20 on @xmath21 . here
@xmath22 is the standard surface measure on @xmath21 , and @xmath23 a given function ; for the simplest case @xmath24 the counterpart of would be @xmath25 for all @xmath16 harmonic on a neighbourhood of @xmath26 .
such domains are referred to as quadrature surfaces , with corresponding pde @xmath27 with @xmath28 , where @xmath29 is a normalization constant .
this problem is well - studied in the continuum case and there is a large amount of literature concerning existence , regularity and geometric properties of both solutions and the free boundary , see @xcite , @xcite , and the references therein . in our search
to find a model for the sandpile particle dynamics to model we came across a few models that we ( naively ) thought could and should be the answer to our quest .
however , running numerical simulations ( without any theoretical attempts ) it became apparent that the models we suggested are far from the bernoulli type free boundaries admitting quadrature identity , in continuum case , should be a sphere . ] .
in particular , redistribution of an initial point mass does not converge ( numerically ) to a sphere , but to a shape with a boundary remotely resembling the boundary of the abelian sandpile ( see figure [ fig-1 ] ; cf .
* figure 1 ) , or ( * ? ? ? * figure 2 ) for instance ) .
[ fig : sub1 ] [ fig : sub2 ] [ fig : sub1 ] [ fig : sub2 ] notwithstanding this , our model seems to present a new alluring and fascinating phenomenon not considered earlier , neither by combinatorics nor free boundary communities . hence the birth " of this article .
since the boundary of a set will be a prime object in the analysis of the paper , we will assume throughout the text that @xmath1 to avoid uninteresting cases .
given an initial mass @xmath2 on @xmath0 , we want to find a model that builds a domain @xmath30 such that the corresponding ( discrete ) balayage measure @xmath4 on the boundary @xmath21 ( see ) is prescribed .
such a model seems for the moment infeasible for us .
a slight variation of it asks to find a
canonical " domain such that the boundary mass @xmath4 stays below the prescribed threshold .
since larger domains ( with reasonable geometry ) have larger boundaries , and hence a greater amount of boundary points , we should expect that one can always find a solution to our problem by taking a very large domain containing the initial mass .
this , in particular , suggests that a canonical domain should be the smallest domain among all domains with the property that the boundary mass @xmath4 is below the prescribed mass , which we shall assume to be uniform mass .
results in a so - called bernoulli free boundary problem , where the green s potential @xmath31 of the mass @xmath2 , in the sought domain @xmath3 , has boundary gradient @xmath32 , in case of uniform distribution , i.e. @xmath33 . ] departing from the above point of view of canonicity one may ask whether there is a natural lattice growth model which corresponds to this minimality . with @xmath34 . ]
it turns out that there actually is a model which seemingly is more complicated than the divisible sandpile model described above . to introduce our model , which we call _ boundary sandpile _
, we need a few formal definitions ) , we think it might be more convenient without any further specification with descriptive names to use the term boundary sandpile . ] .
through the text various letters @xmath35 will stand for constants which may change from formulae to formula .
we use lower case @xmath36 to denote small constants , and upper case @xmath37 for large constants . call two lattice points @xmath38 _ neighbours _ and write @xmath39 if @xmath40 , where @xmath41-norm of @xmath42 is the sum of absolute values of its coordinates .
clearly @xmath39 iff @xmath43 for some @xmath44 where @xmath45 is the @xmath46-th vector of the standard basis of @xmath7 . for any non - empty set @xmath47
, this concept of lattice adjacency induces a natural graph @xmath48 having vertices on @xmath49 .
we call such @xmath48 the _ lattice graph _ of @xmath49
. next , for a set @xmath47 define @xmath50 and call them respectively the * boundary * of @xmath49 and the * interior * of @xmath49 . for the convenience of the exposition we have chosen to include the boundary into the set . for @xmath51 recall the definition of the _ discrete laplace operator _ denoted by @xmath52 and acting on a function @xmath53 by @xmath54 , \qquad x\in h{{\mathbb z}}^d,\ ] ] where @xmath55 , as for @xmath56 , means that @xmath57 is a neighbour of @xmath58 in the lattice @xmath59 . throughout the text , when @xmath56 , i.e. on a unit scale lattice , we will write @xmath60 for @xmath61 unless explicitly stated otherwise
. for @xmath62 we set @xmath63 for the closed discrete ball of radius @xmath64 including its 1-neighbourhood .
it is clear from the definition that @xmath65 is from the interior of @xmath66 iff @xmath67 .
let now @xmath68 be a given ( mass ) distribution on @xmath0 , i.e. a non - negative function supported on finitely many sites of @xmath0 .
fix also a threshold @xmath69 . for each integer
@xmath70 we inductively construct a sequence of sets @xmath71 , mass distributions @xmath72 , and functions @xmath73 as follows .
start with @xmath74 and @xmath75 .
for an integer time @xmath70 a particular site @xmath65 is called * unstable * if either of the following holds : * @xmath76 and @xmath77 , * @xmath78 and @xmath79 .
otherwise a site is called * stable*. we call the number @xmath80 the * ( boundary ) capacity * of the model and refer to @xmath71 as the set of _ visited sites _ at time @xmath81 .
any unstable site can _ topple _ by distributing all its mass equally among its @xmath82 lattice neighbours .
more precisely , for each @xmath70 we choose an unstable site @xmath83 and define @xmath84 , @xmath85 and @xmath86 , @xmath87 , where @xmath88 is the kronecker delta symbol at the origin , i.e. @xmath89 equals 1 if @xmath90 and is zero otherwise for @xmath65 .
we call @xmath73 the _ odometer _ function at time @xmath81 . for the sake of convenience
, we do allow the toppling to be applied to a stable site , as an identity operator , i.e. if at time @xmath81 a toppling is applied to a stable site @xmath58 , then we set @xmath91 , @xmath92 and @xmath93 .
we say that toppling @xmath58 is _ legal _ , if @xmath58 is unstable .
if for some @xmath81 there are no unstable sites , the process is terminated .
we call this model * boundary sandpile * ( @xmath94 ) and denote by @xmath95 , where @xmath68 is the initial distribution , and @xmath80 is the boundary capacity of the model . it is clear that the triple @xmath96 may depend on the choice of the unstable sites , i.e. the toppling sequence . later on
we will see that for a suitable class of toppling sequences stable configurations exist and are identical ( see propositions [ prop - stabil - toppling ] and [ prop - abelian ] ) .
observe that from the definition of discrete laplacian and ( [ mu - k+1-def ] ) we easily see that for each @xmath70 one has @xmath97 i.e. the laplacian of @xmath73 represents the net gain of mass for a site @xmath58 at time @xmath81 .
we further define a few concepts needed for our analysis . for a @xmath95 , and a ( toppling ) sequence @xmath98 , we say that @xmath99 is * stabilizing * if there exists a distribution @xmath2 such that @xmath100 as @xmath101 for any @xmath65 .
we call @xmath99 * infinitive * on a set @xmath47 if every @xmath102 appears in the sequence @xmath99 infinitely often . if the set on which @xmath99 is infinitive is not specified , then it is assumed to be the entire @xmath0
. our main results concern general qualitative analysis of the boundary sandpile model , introduced in this paper . for any initial mass distribution
we prove well - posedness of the model ( propositions [ prop - stabil - toppling ] and [ prop - abelian ] ) , and canonical representation of the model in terms of the smallest super - solution among a certain class of functions ( theorem [ thm - canonical ] ) . specifying our analysis for point masses and using this minimality of the model ,
we show directional monotonicity of the odometer function ( theorem [ thm - monotonicity ] ) .
we then determine the reasonable size of the boundary capacity ( subsection [ sub - sec - heuristics ] ) and estimate the growth rate of the model in lemma [ lem - balls - inside - out ] .
we prove a uniform lipschitz bound for the scaled odometer function in proposition [ lem - lipschitz - est ] . using these results obtained in discrete setting , in the final part of the paper , section [ sec - shape - analysis ] we study the scaling limit of the model in continuum space .
in particular , we show that ( along a subsequence ) the scaled odometers , and hence the visited sites , converge to a continuum limit ( theorem [ thm - scaling - limit ] parts ( i ) and ( iii ) ) .
we also prove that the free boundary of ( any ) scaling limit of the model is locally a lipschitz graph ( theorem [ thm - scaling - limit ] part ( v ) ) .
in subsection [ subsec - asm ] we apply our methods developed for boundary sandpile model to classical abelian sandpile .
most notably , we show that the boundary of the scaling limit of abelian sandpile corresponding to initial configuration of chips at a single vertex , has lipschitz boundary .
[ rem - general - f ] it is worthwhile to remark that the boundary sandpile described above can be represented slightly different , using discrete partial derivatives . indeed , our model generates a set @xmath47 whose green s function @xmath10 ( corresponding to the given initial distribution of mass ) satisfies @xmath103 for a given @xmath104 , where @xmath105 . in light of this observation one may consider a wider class of problems in terms of general prescribed boundary mass given by @xmath106 , where @xmath107 defined on @xmath108 has to satisfy some properties ( ellipticity and other ) , yet to be found .
the methods developed in this paper seem to have a good chance to go through for at least nice enough " function @xmath107 .
this aspect we shall leave for interested reader to explore .
in this section we prove two basic properties of the boundary sandpile model .
namely , that the model needs to visit a finite number of sites in @xmath0 to reach a stable state , and that the final stable configuration is independent of the toppling sequence . for the proofs we will use some ideas from (
* lemma 3.1 ) and ( * ? ? ?
* lemma 3.1 ) .
[ prop - stabil - toppling ] for a given @xmath95 any infinitive toppling sequence @xmath99 is stabilizing . let @xmath109 be the total mass of the sandpile , and set @xmath110 . for each @xmath111 let @xmath71 be the set of visited sites after invoking @xmath81-th toppling in @xmath99 .
let also @xmath73 and @xmath72 be respectively the odometer function and the distribution of mass after the @xmath81-th site in @xmath99 has toppled .
we first show that for any @xmath111 and for each @xmath76 satisfying @xmath112 one has @xmath113 indeed , let @xmath114 be the smallest integer such that @xmath115 , and let @xmath116 be the neighbour of @xmath58 that topples at time @xmath46 . since @xmath116 and @xmath117 we have @xmath118 , but as toppling of @xmath57 is legal ( since it is producing a previously unvisited site @xmath58 ) , we must have @xmath119 .
consequently we get @xmath120 , and since @xmath58 did not topple at steps @xmath121 , as otherwise @xmath58 will be in the interior of @xmath71 , we get @xmath122 and hence ( [ mu - k - from - below ] ) . comparing ( [ mu - k - from - below ] ) with the total mass @xmath123 , for any @xmath70 we get latexmath:[\[\label{bound - on - bdry - points } upper bound for the number of newly emerged .
the simple reason is that in general @xmath68 may have , say , arbitrarily large number of isolated points with small mass , which may never be visited again in the course of the process . ]
boundary points of visited sites at any time of the process . now assume that @xmath71 is connected as a lattice graph .
fix @xmath125 and let @xmath126 be such that @xmath127 and @xmath128 . for each integer @xmath129 consider the hyperplane @xmath130 .
it is clear that @xmath131 contains at least one element of @xmath132 and these points are pairwise different for integer @xmath133 $ ] . by ( [ bound - on - bdry - points ] )
one has @xmath134 as @xmath125 is arbitrary , for any @xmath83 we obtain @xmath135 hence the process occupies a finite region of @xmath0 independently of we will obtain more precise bounds for the set of visited sites .
however , the bound in lemma [ lem - balls - inside - out ] uses a priori boundedness of the set of visited sites . ]
it is clear that assuming connectedness of @xmath71 resulted in no loss of generality , as we can employ the scanning procedure by hyperplanes on each connected component of the lattice graph of @xmath71 .
since the toppling is infinitive , for each @xmath136 all unstable sites on @xmath132 will topple eventually , by so extending the set of visited sites .
but ( [ vis - sites - bound ] ) shows that after finite number of topplings , the boundary of visited sites will become stable and will remain so throughout . precisely , there is @xmath136 and a set @xmath47 such that @xmath137 for all @xmath138 . by ( [ laplace - of - odometer ] ) for any integer @xmath70 we have @xmath139 the _ r.h.s .
_ of the last expression , as well as each @xmath73 are supported in @xmath49 for all integer @xmath70 .
we thus have that the functions @xmath140 , independently of @xmath111 , are uniformly bounded on @xmath0 , and have compact supports .
take any @xmath141 , then @xmath142 and @xmath143 in view of the stability of @xmath49 .
hence @xmath144 which shows that @xmath145 is uniformly bounded for any @xmath146 .
this uniform bound on @xmath73 propagates into each connected component of the lattice graph of @xmath49 .
hence we get @xmath147 for any @xmath65 and any @xmath136 with some constant @xmath148 . since for each @xmath65 , the sequence @xmath149 is increasing and bounded above
, it has a finite limit which we denote by @xmath150 .
clearly @xmath10 is non - negative and compactly supported .
it also follows that the _ l.h.s . _ of ( [ u - k - laplacian ] ) must have a limit everywhere in @xmath0 . passing to the limit in ( [ u - k - laplacian ] )
as @xmath81 goes to infinity , we get @xmath151 in @xmath0 where by @xmath2 we denote the limit distribution of @xmath72 . to complete the proof it is left to observe that the final distribution @xmath2 is stable .
as we saw above , @xmath2 is stable on @xmath152 , hence it remains to show that @xmath153 in the interior of @xmath49 .
fix any @xmath154 , since @xmath99 is infinitive we get @xmath155 for infinitely many @xmath81 . but as @xmath156 and the limit exists , we obtain @xmath157 completing the proof of the proposition .
it is clear from the proof of proposition [ prop - stabil - toppling ] that one can require the toppling to be infinitive only on the set of visited sites .
we do not do so as the final domain is not known a priori .
we call the function @xmath10 constructed in the proof of proposition [ prop - stabil - toppling ] the _ odometer _ and the distribution @xmath2 the _ final configuration _ corresponding to the given toppling sequence .
[ prop - abelian ] for a given @xmath95 and any two infinitive toppling sequences @xmath158 and @xmath159 , one has @xmath160 for the corresponding odometer functions . as in proposition
[ prop - stabil - toppling ] by @xmath123 denote the total mass of the system . in view of proposition
[ prop - stabil - toppling ] each @xmath161 is stabilizing and has well - defined odometer function , call it @xmath162 .
we will prove that @xmath163 for any @xmath65 , which by symmetry will imply the desired result .
for @xmath164 by @xmath165 denote the odometer function and by @xmath166 the distribution corresponding to the sequence @xmath161 after the @xmath81-th toppling occurs .
let @xmath167 , we now show that @xmath168 observe that ( [ ineq - abelian ] ) is enough for our purpose since the site @xmath169 appears infinitely often in @xmath158 and @xmath170 converges to @xmath171 pointwise as @xmath81 tends to infinity .
thus , in what follows we prove ( [ ineq - abelian ] ) which we will do by induction on @xmath81 . when @xmath172 , ( [ ineq - abelian ] ) is trivially true .
now assume that it holds for any time @xmath173 .
we divide the analysis into two cases . * case 1 . * _ toppling @xmath169 in @xmath158 at time @xmath81 is not a legal move_. in this case we have @xmath174 so , if @xmath169 has never toppled prior to time @xmath81 , then @xmath175 and we are done . otherwise , let @xmath176 be the last time @xmath169 has toppled in @xmath158 .
we thus have @xmath177 , hence , using the inductive hypothesis , we conclude @xmath178 where the last equality follows from ( [ non - legal ] ) .
this completes the induction step for the case of a non - legal move .
* _ toppling @xmath169 in @xmath158 at time @xmath81 is legal_. we first show that for any @xmath179 one has @xmath180 there are two possible sub - cases , either @xmath58 was never toppled in @xmath158 up to time @xmath81 , which implies @xmath181 and we get ( [ u2-geq - u1k ] ) , or @xmath58 was toppled at some time before @xmath81 . in the latter case let @xmath182 be the last time @xmath58 has toppled prior to time @xmath81 .
observe that @xmath183 since @xmath179 .
then we have @xmath184 where the second inequality follows by inductive hypothesis .
we thus have proved ( [ u2-geq - u1k ] ) for all @xmath179 .
consider the following inequality @xmath185 to complete the induction step suppose for a moment that ( [ mu2-dist - vs - mu1k ] ) holds true . then @xmath186 rearranging the first and the last terms leads to @xmath187 where the last inequality is due to ( [ u2-geq - u1k ] ) .
this completes the induction .
thereby , to finish the proof of the proposition , we need to verify ( [ mu2-dist - vs - mu1k ] ) which we do next . to prove ( [ mu2-dist - vs - mu1k ] ) we will assume that @xmath169 is on the boundary of the visited sites generated by @xmath159 , call it @xmath188 , as otherwise the _ l.h.s .
_ of ( [ mu2-dist - vs - mu1k ] ) vanishes and the inequality is trivial .
recall , that we are in case 2 , hence there are two possible reasons for @xmath169 to topple in @xmath158 at time @xmath81 .
namely , * @xmath169 was on the boundary of visited sites of @xmath158 at time @xmath189 and had mass strictly greater than @xmath80 , * @xmath169 was in the interior of visited sites of @xmath158 at time @xmath189 and had positive mass .
we show that either of the cases leads to a contradiction .
assume ( a ) , then @xmath190 which is a contradiction .
now consider the case when @xmath169 topples according to ( b ) .
recall that @xmath191 .
this implies that @xmath169 can not topple in @xmath158 prior to time @xmath81 .
indeed , if @xmath169 topples for some @xmath183 , then @xmath192 , but by inductive hypothesis we have @xmath193 which is false .
next let @xmath183 be the first time when the toppling of @xmath158 produces a vertex outside @xmath194 , and let @xmath57 be the site that topples at time @xmath46
. on one hand @xmath57 has to be on the boundary of the set visited by @xmath159 , on the other hand , as we saw already , @xmath57 must be different from @xmath169 .
hence by ( [ u2-geq - u1k ] ) and the fact that @xmath195 we get @xmath196 meaning that @xmath57 can not topple .
we arrive at a contradiction , hence ( [ mu2-dist - vs - mu1k ] ) is proved , and so is the proposition .
a simple consequence of the abelian property is the following .
[ cor - o - symm ] let the initial distribution @xmath68 be a point mass supported at the origin , and let @xmath49 be its final domain of visited sites , and @xmath10 be the odometer function
. then @xmath49 is symmetric with respect to coordinate axes , and @xmath197 for all @xmath65 . for each @xmath198 consider the class @xmath199 : = \ { ( { \varepsilon}_1 x_1, ... ,{\varepsilon}_d x_d ): \ { \varepsilon}_i\in \{-1,1\ } , \
i=1, ... ,d \}.\ ] ] take any infinitive toppling @xmath200 for @xmath68 and consider its symmetrized " version @xmath201 = \ { [ x^{(k ) } ] \}_{k=1}^\infty$ ] , where each @xmath169 in @xmath99 is replaced by its equivalence class in any ordering .
clearly the new toppling sequence @xmath202 $ ] is infinitive , and hence stabilizing in view of proposition [ prop - stabil - toppling ] .
also , if @xmath73 is the odometer function after toppling the entire class @xmath203 $ ] in @xmath202 $ ] , then it is axially symmetric by construction , namely @xmath204 for any @xmath205 $ ] .
since the odometer @xmath10 corresponding to @xmath202 $ ] is the pointwise limit of @xmath73 , the claim follows .
one can see , in a similar way , that for any initial distribution @xmath68 , its final configuration @xmath49 and odometer function @xmath10 inherit the symmetries of @xmath68 .
we do not formulate this rigorously , as later on we will be using the symmetry for point masses only .
it is apparent that our boundary sandpile process requires infinite number of toppling steps to reach a stable configuration ( see remark [ rem - infinite - steps ] below ) .
nonetheless , it is not hard to observe that the process can be stabilized effectively in finite number of steps by absorbing part of the topplings into certain linear operators acting on graphs .
this fact might also be useful for numerical simulations .
let @xmath206 be a finite connected , simple ( i.e. without loops and multiple edges ) graph .
partition @xmath49 into two non - empty subsets @xmath207 and @xmath208 which we call * source * and * sink * respectively . for a given function @xmath209 , in analogy with ( [ mu - k+1-def ] ) ,
consider the toppling transformation @xmath99 acting on a vertex @xmath210 by @xmath211 where @xmath212 is the degree of the vertex @xmath213 in @xmath48 , which is always positive in view of the connectedness of the graph .
we also denote @xmath214 = \frac{1}{|v_0|}\sum\limits_{v\in v_0 } \mu(v ) $ ] for the average of @xmath2 . with these preparations
we easily obtain the following .
[ lem - graph - balayage ] for any map @xmath215 such that the pre - image of each @xmath210 is infinite , for all @xmath210 we have @xmath216 we will assume without loss of generality that the subgraph on @xmath207 is connected , as otherwise we can apply the argument that follows to each connected component of the subgraph .
set @xmath217 and let @xmath218 be the diameter of the subgraph on @xmath207 , i.e. the maximal length of the shortest path between any two vertices .
assume that the maximum of @xmath2 is attained at some @xmath210 , and take any shortest path from @xmath213 to the sink @xmath208 through the vertices of @xmath207 .
let @xmath219 be such path where each @xmath220 .
we have @xmath221 by definition . from the definition of @xmath222
we can choose @xmath223 large enough such that the sequence @xmath224 is a subsequence of @xmath225 .
but after toppling @xmath226 the total mass of @xmath207 is decreased by at least @xmath227 , and as @xmath228 , we have @xmath229 \leq \mathbb{e}[\mu ] \left(1- \frac{1}{|v_0| } \frac{1}{d_\ast^\rho } \right).\ ] ] since the average is reduced by a constant multiple the claim follows .
[ rem - abelian - graph ] by adapting the proof of proposition [ prop - abelian ] ( in fact in a simpler form ) one can easily show that the amount of mass each vertex of the sink receives is independent of the toppling order , as long as each source vertex is toppled infinite number of times .
[ rem - infinite - steps ] take a graph with two vertices and a single edge connecting them .
let also each of these two vertices be attached to any number of sinks ( possibly none , but not simultaneously ) .
then clearly one has to topple each vertex infinitely often in order to entirely transfer any given mass to the sinks .
hence the requirement on each vertex to topple infinite number of times can not be eliminated .
we will see shortly that one can avoid a straightforward application of topplings to transform mass from _ source _ to _
sink_. by realizing this transference as a linear operator we will perform the mass transportation from source to sink effectively in one step . keeping the notation of lemma [ lem - graph - balayage ]
let the source be @xmath230 and the mass vector be equal to @xmath231 .
consider the toppling @xmath232 and for @xmath111 let @xmath233 denote the @xmath81-th iterate of @xmath99 .
then by lemma [ lem - graph - balayage ] we have that @xmath234 as @xmath235 and the goal is to determine how much mass each sink vertex will receive in the limit . due to the abelian property ( remark [ rem - abelian - graph ] )
the amount of mass distributed to the sinks is independent of the toppling sequence , hence specifying the toppling order results in no loss of generality . to compute the final distribution of the mass on the sink
we introduce matrices @xmath236 where we call @xmath237 the * mass transformation * matrix , and @xmath3 the * mass distribution * matrix .
more precisely @xmath238 shows the proportion of @xmath239 which is present at vertex @xmath240 after we have applied the transformation @xmath99 to the graph .
namely if @xmath241 is the new mass - vector after applying @xmath99 , then @xmath242 ( v_i ) = m_{i1 } \mu_0(v_1 ) + m_{i2 } \mu_0(v_2)+ ... +m_{i n } \mu_0(v_n).\ ] ] in a similar fashion @xmath243 shows the proportion of @xmath239 transferred to its neighbours by @xmath240 .
thus each sink ( if any ) attached to @xmath240 gets mass equal to @xmath244 a simple bookkeeping procedure illustrated below in algorithm [ algo-1 ] allows us to compute these coefficients when we consequently topple @xmath245 .
* input : * connected graph @xmath206 on source vertices @xmath246 initialize @xmath237 to @xmath247 _ identity _ and @xmath3 to @xmath247 _ zero _ matrices . @xmath248
@xmath249 @xmath250 * output : * matrices @xmath237 and @xmath3 [ algo-1 ] now observe that after the toppling @xmath233 , @xmath111 , is applied , the mass vector distributed to sinks becomes @xmath251+ ...
+ d[m^{k-1 } \mu_0 ] = d \left [ \sum\limits_{i=0}^{k-1 } m^i \right ] \mu_0 \in { { \mathbb r}}^n.\ ] ] to sum the series in ( [ neumann - series ] ) we need the following .
[ lem - e - value1 ] all eigenvalues of @xmath237 are less than 1 by absolute value .
take any vector @xmath252 with all coordinates being non - negative . by virtue of ( [ exp - decay ] )
there exists an integer @xmath136 and a constant @xmath253 such that @xmath254 \leq ( 1-c ) \mathbb{e } [ \mu_0].\ ] ] since @xmath237 has non - negative entries due to construction by algorithm [ algo-1 ] , from the last inequality we get @xmath255 now for any @xmath256 .
hence , if @xmath257 is an eigenvector of @xmath237 corresponding to an eigenvalue @xmath258 , we get @xmath259 which implies @xmath260 completing the proof of the lemma .
lemma [ lem - e - value1 ] implies that the matrix @xmath261 is invertible , where @xmath262 is the @xmath263 identity matrix .
hence the neumann series of @xmath237 , as we have in ( [ neumann - series ] ) , is convergent . passing to limit in ( [ neumann - series ] ) as @xmath264 demonstrates that the total mass @xmath2 distributed to the sinks equals @xmath265 \mu_0 = d ( \mathrm{id}- m)^{-1 } \mu_0.\ ] ] observe that all mass is being transferred to the sink by lemma [ lem - graph - balayage ] , and hence there is no loss of mass .
this shows that the linear operator defined by ( [ mass - trans - sink ] ) is non - degenerate .
hence , the matrix @xmath3 is invertible as well .
finally , each sink vertex attached to @xmath240 gets mass equal to the @xmath46-th component of the vector @xmath266\in { { \mathbb r}}^n.\ ] ] based on the above discussion , we next present a simple pseudocode , which allows to stabilize the given initial distribution in finite number of steps .
* input : * initial distribution @xmath68 and boundary capacity @xmath80 initialize @xmath49 , the set of visited sites , as the support of @xmath68 populate a stack @xmath267 of sites from @xmath49 that need to be toppled .
topple(v ) @xmath213 is the top element of the stack include @xmath268 in @xmath267 if not there already remove @xmath213 from @xmath267 construct @xmath48 , the lattice graph on vertices of @xmath269 construct the distribution matrices @xmath237 and @xmath3 for @xmath48 by algorithm [ algo-1 ] use @xmath237 and @xmath3 to distribute the mass from @xmath269 to @xmath132 by re - populate the stack @xmath267 by elements of @xmath152 that need to be toppled * output : * final set of visited sites and stable mass distribution on the boundary [ algo-2 ] in algorithm [ algo-2 ] we simply move the boundary of the set of visited sites as long as there is a possibility .
then , we transfer at once all mass from the interior onto the boundary which requires a passage to the limit in a toppling procedure . consequently , the fact that this will produce the same final configuration as the original boundary sandpile process , is not a direct corollary of abelian property of the sandpile , but follows from lemma [ lem - interim ] .
in particular , due to lemma [ lem - interim ] each iteration of the algorithm resuming from step 3 , starts with an initial distribution @xmath2 for which the two sandpiles @xmath270 and @xmath95 produce the same final configurations .
this fact , coupled with proposition [ prop - stabil - toppling ] implies that algorithm [ algo-2 ] terminates in finite number of steps since the set of visited sites can not grow to infinity .
depending on @xmath68 and @xmath80 , however , it is interesting to observe , that the same approach can stabilize the divisible sandpile of levine - peres @xcite in finite number of steps . ]
we start with the following . * discrete maximum principle ( dmp ) : * let @xmath271 , @xmath47 be finite , and @xmath272 in the interior of @xmath49
. then @xmath273 .
the proof of this important statement is an easy exercise ( see e.g. ( * ? ? ?
* exercise 1.4.7 ) ) .
we now show that the boundary sandpile requires the smallest possible domain in the discrete space , where it can stabilize itself .
[ def - stab - pair ] for an initial distribution @xmath68 and a boundary capacity @xmath80 , we say that the pair @xmath274 is stabilizing for @xmath95 if the following holds : * @xmath47 is a finite set @xmath275 , * @xmath276 is a function satisfying @xmath277 in this setting , we will call @xmath10 odometer function .
[ rem - stabilizing ] for any @xmath68 and @xmath80 let @xmath207 be the set of visited sites of the corresponding @xmath94 and @xmath278 be the odometer function .
then clearly the pair @xmath279 is stabilizing in a sense of definition [ def - stab - pair ] .
[ thm - canonical ] let @xmath280 be the set of visited sites of @xmath281 .
then @xmath207 is the intersection of all @xmath47 for which there is a function @xmath10 such that the pair @xmath282 is stabilizing for @xmath95 .
the set of stabilizing pairs is non - empty in view of remark [ rem - stabilizing ] . hence if @xmath283 is the intersection in the formulation of the theorem , then @xmath283 is finite and contains the support of @xmath68 .
first , we show that for some @xmath284 the pair @xmath285 is stabilizing . to see this for @xmath283 ,
it is enough to prove that for any two stabilizing pairs @xmath286 , @xmath164 the set @xmath287 has the property . to this end , consider the function @xmath288 in @xmath49 .
clearly @xmath289 from which we get @xmath290 on @xmath152 .
it is also clear that @xmath291 in @xmath292 .
next , using the fact that @xmath213 vanishes on @xmath152 and is the pointwise minimum of @xmath171 and @xmath293 , from ( [ bdry - as - union ] ) we have @xmath294 now let @xmath268 be the unique solution to @xmath295 since @xmath296 is a subsolution in the interior of @xmath49 , from dmp we have @xmath297 in @xmath49 . combining this with the fact that both @xmath268 and @xmath213 vanish on @xmath152 we arrive at @xmath298 but then , @xmath299 where the last inequality is due to ( [ v0-on - v0 ] )
hence , the pair @xmath300 is stabilizing and the claim for @xmath283 follows .
we are now in a position to show @xmath301 . by definition of @xmath283
we have @xmath302 .
take any toppling sequence @xmath303 where @xmath304 for all @xmath305 and @xmath99 is infinitive might be empty , in which case @xmath99 is empty as well , that is no toppling occurs .
this corresponds to the case , when the initial distribution @xmath68 is already stable . ] on @xmath306 .
observe that if @xmath99 is stabilizing for the original @xmath94 , then @xmath307 as the final configuration is unique by abelian property of the sandpile .
hence , it is left to prove that @xmath99 is stabilizing .
assume it is not .
invoking @xmath99 , we get that all mass from the interior of @xmath283 is being redistributed onto @xmath308 by lemma [ lem - graph - balayage ] .
we have @xmath309 , and since @xmath99 is not stabilizing it follows that the distribution of mass on @xmath283 obtained after applying @xmath99 is not stable .
since no mass remains in the interior of @xmath283 it follows that the boundary of @xmath283 is unstable .
let @xmath10 be the odometer function corresponding to the toppling @xmath99 .
then , by definition we have @xmath310 let also @xmath284 be the odometer for which @xmath311 is stabilizing , whose existence was established above . by definition ,
@xmath284 satisfies ( [ u - in - int - v - star ] ) hence , due to uniqueness of solutions , we get that @xmath312 . but then
@xmath313 where the inequality follows since the pair @xmath285 is stabilizing . this shows , that @xmath314 is stabilizing as well , which contradicts our assumption that the toppling @xmath99 was not stabilizing , and completes the proof of the theorem .
[ def - stab - pair ] for a given @xmath95 we call @xmath274 _ the stabilizing pair _ of the sandpile if @xmath49 is minimal in a sense of theorem [ thm - canonical ] .
an immediate reformulation of the previous theorem is the following characterization of the odometer as the smallest supersolution in a certain class of functions on @xmath0 . for @xmath95 let @xmath10 be the odometer function .
then @xmath315 where @xmath316 .
a useful corollary of the minimality principle , which is being used in algorithm [ algo-2 ] above , is the following .
[ lem - interim ] let @xmath317 be the stabilizing pair for @xmath95 . fix any @xmath318 such that @xmath319 and let @xmath320 be the solution to @xmath321 then the final configurations of @xmath95 and @xmath322 are identical
. the lemma states that if we balayage @xmath68 to @xmath323 first , and then balayage the resulting measure from @xmath323 to @xmath324 , the boundary measure produced on @xmath324 will be the same as if we had balayaged @xmath68 on @xmath324 .
this is obvious from a potential theoretic and balayage perspective .
it is however not straightforward from a @xmath94 dynamic point of view .
set @xmath325 , which , by the definition of @xmath320 , is a distribution supported on @xmath323 .
let @xmath326 be the stabilizing pair for @xmath327 .
the aim is to show that @xmath328 and that the final configurations for @xmath95 and @xmath327 coincide .
the latter is equivalent to equality @xmath329 on @xmath0 , which in turn is equivalent to @xmath330 which we will prove below .
we have @xmath331 as well as @xmath332 on @xmath333 .
clearly @xmath334 therefore @xmath335 on @xmath333 .
we also have @xmath336 hence the pair @xmath337 is stabilizing for @xmath281 . as a consequence of theorem [ thm - canonical ] we obtain @xmath338 .
we now prove the reverse inclusion .
since @xmath318 , it follows that @xmath339 is supported in @xmath207 .
hence , we may define @xmath340 as the unique solution of @xmath341 since @xmath342 it follows that @xmath343 on @xmath324 .
this implies that the function @xmath344 is vanishing on @xmath324 .
but we also have @xmath345 and by the uniqueness of solutions to the dirichlet problem , we infer that @xmath346 in @xmath207 .
in particular it follows that @xmath347 which , by the stability of @xmath278 , implies that @xmath348 is stabilizing for @xmath327 .
again , by the minimality principle we get @xmath349 , consequently @xmath328 .
it is left to show ( [ config - equiv ] ) . since @xmath328 ,
the function @xmath350 vanishes on @xmath324 and is identically 0 outside @xmath207 .
hence in the interior of @xmath207 we have @xmath351 which shows that @xmath352 on @xmath0 and hence ( [ config - equiv ] ) .
the lemma is proved .
yet another important consequence of the minimality principle of theorem [ thm - canonical ] is a special directional monotonicity of the odometer function corresponding to an initial distribution concentrated at a single point .
this key property of the boundary sandpile will prove crucial in studying the regularity of the boundary of scaling limit of the process .
let @xmath353 be the set of mirror symmetry hyperplanes of the unit cube @xmath354^d$ ] .
more precisely , @xmath353 consists of the hyperplanes @xmath355 , @xmath356 and @xmath357 where @xmath358 with @xmath359 .
in particular , @xmath353 has @xmath360 elements in total .
[ thm - monotonicity ] assume @xmath282 is the stabilizing pair of @xmath361 where @xmath362 and @xmath69 .
fix any hyperplane @xmath363 . then
, for any @xmath364 , such that @xmath365 is a non - zero vector orthogonal to @xmath99 , we have @xmath366 this result formalizes the intuition that moving away from the origin , where the total mass of the sandpile was concentrated initially , emissions of mass should become smaller .
there are however restrictions , partially due to our proof , on the way one moves on the lattice . namely the theorem allows to compare points lying on a lattice - line " .
the proof is motivated by a.d .
alexandrov s celebrated _ moving plane _
method @xcite , see also @xcite .
we will assume @xmath367 and will show that @xmath368 .
this clearly implies the claim of the theorem . there is a unique translated copy of @xmath99 ,
call it @xmath369 , having equal distance from @xmath370 and @xmath371 . for a given @xmath42 denote by @xmath372 its mirror reflection with respect to the hyperplane @xmath369 . due to the choice of @xmath99 and @xmath370 ,
@xmath371 it follows that each lattice point is being reflected in this way to a lattice point , and since reflection is an involution , we have that @xmath373 , in particular @xmath374 .
set @xmath375 for the reflection of @xmath49 , and similarly define the reflected odometer function by @xmath376 , where @xmath65 . from the discussion above , we have that @xmath377 is defined on @xmath0 and @xmath378 .
let @xmath379 be the closed halfspace of @xmath7 determined by @xmath369 and containing the origin , and let @xmath380 .
consider the set @xmath381 where @xmath382 and @xmath383 .
our first goal is to show that @xmath384 . by definition @xmath385 , so we need to establish the reverse inclusion .
consider the function @xmath386 we claim that @xmath387 indeed , fix @xmath58 in the interior of @xmath388 .
there are two possibilities
. * case 1 .
* @xmath389 . in this case ( [ laplace - u - t ] ) follows from @xmath390 * case 2 . *
@xmath391 .
we have two sub - cases here , namely either @xmath392 or @xmath393 . in the former case , ( [ laplace - u - t ] )
follows as in case 1 above .
we will therefore assume that @xmath394 then we have @xmath395 observe , that if the first sum on the last row of ( [ u - t - case2 ] ) is empty , then we get @xmath396 . otherwise , there are two possible scenarios treated below .
the first one occurs when the hyperplane @xmath99 is orthogonal to a coordinate axis , i.e. @xmath99 is one of @xmath397 for some @xmath44 . by assumption
there is @xmath398 such that @xmath399 .
but due to the choice of @xmath99 this is possible if and only if @xmath400 and @xmath401 .
in particular , the above - mentioned sum contains a single element , namely the one arising from @xmath400 . taking into account ( [ u - t - as - min ] ) ,
from we obtain @xmath396 .
consider the second scenario , when the symmetry hyperplane @xmath99 is not orthogonal to any of the coordinate axes , and assume that there exists @xmath398 such that @xmath399 . here
as well , in view of the choice of @xmath99 , such @xmath57 must lie on the hyperplane @xmath369 , and hence @xmath402 .
thus the first sum in the last row of covers all lattice neighbours of @xmath372 lying on @xmath369 .
observe that all lattice neighbours of @xmath372 are in @xmath403 , so they are either on @xmath369 , or are obtained as reflections of lattice neighbours of @xmath58 from @xmath404 .
the first class of neighbours is covered by the first sum , and the second class by the second sum of of the last row of .
hence , the last row of equals @xmath405 .
we conclude that in case 2 @xmath406 finally , notice that for @xmath391 lying in the interior of @xmath388 we have that @xmath372 is from the interior of @xmath388 .
hence , @xmath407 since the origin is in @xmath403 .
we thus get from .
next , we proceed to the final stage of the proof , where we show that @xmath408 everywhere on @xmath0 .
it is easy to observe that @xmath409 on @xmath410 and @xmath411 on @xmath410 .
now let @xmath213 be the solution to @xmath412 in the interior of @xmath388 and @xmath290 on @xmath410 .
by dmp we get @xmath413 everywhere on @xmath388 , hence @xmath414 on @xmath410 as both @xmath213 and @xmath415 vanish on the boundary of @xmath388 .
it follows that the pair @xmath416 is stabilizing for @xmath361 ( notice that the origin lies in @xmath388 ) , hence the minimality principle due to theorem [ thm - canonical ] in conjunction with the inclusion @xmath385 implies @xmath384 .
in particular we get @xmath408 everywhere on @xmath49 since the difference @xmath417 defines a supersolution in @xmath49 with zero boundary data .
lastly , since both @xmath10 and @xmath415 vanish outside @xmath49 , we get that @xmath408 on @xmath0 . from here
, the desired inequality follows easily .
indeed , since @xmath418 and the origin lies in @xmath403 , we must have @xmath419 . due to construction
we also have @xmath374 , hence @xmath420 the proof of the theorem is completed .
the main aim of this subsection is the analysis of the well - known _ abelian sandpile model _
( asm ) via tools developed in this paper .
in particular , we will show that the scaling limit of asm generated from an initial configuration of chips placed at a single vertex has lipschitz boundary .
while some aspects of the local ( inner ) geometry of asm were analysed in @xcite , we believe that this is the first occurrence addressing the geometry of the boundary .
we briefly recall the definition of the process .
asm is a lattice growth model for configurations of chips distributed on vertices of @xmath0 .
a vertex carrying at least @xmath82 chips topples giving a single chip to all its @xmath82 lattice neighbours , and losing @xmath82 chips itself .
if there are no sites with more than @xmath421 chips , the process terminates . for any finite non - negative initial configuration of chips , subsequently toppling all sites with at least @xmath82 chips , the process terminates in finite steps .
this process is abelian in the sense that the final configuration of chips is independent of the order of topplings .
fix @xmath422 , and let @xmath10 be the odometer function for the starting configuration of chips @xmath423 .
that is , for each @xmath65 the value @xmath150 is the number of topplings required to reach stable configuration in asm .
due to abelian property the function @xmath424 is well - defined .
let also @xmath425 be the final ( stable ) configuration of chips .
similarly to ( [ laplace - of - odometer ] ) one can easily see that @xmath426 we will need the following characterization proved by fey , levine , and peres @xcite .
* least action principle*. let @xmath123 and @xmath10 be as above .
then @xmath427 we have the analogue of theorem [ thm - monotonicity ] for asm . for the definition of @xmath353 see the paragraph above theorem [ thm - monotonicity ] .
[ thm - monotonicity - asm ] for a given @xmath422 let @xmath10 be the odometer for initial distribution @xmath423 .
fix any hyperplane @xmath363 . then , for any @xmath364 , such that @xmath365 is a non - zero vector orthogonal to @xmath99 , we have @xmath366 we will only sketch the proof as it is a simplification of the proof of theorem [ thm - monotonicity ] . assuming
@xmath367 will show that @xmath368 , which implies the claim of the theorem .
let @xmath369 be as in the proof of theorem [ thm - monotonicity ] , and for a given @xmath42 denote by @xmath372 its mirror reflection with respect to the hyperplane @xmath369 . recall that we have @xmath374 .
define also the reflected odometer function by @xmath376 , where @xmath65 .
we have that @xmath377 is defined on @xmath0 with values in @xmath428 .
let @xmath379 be the closed halfspace of @xmath7 determined by @xmath369 and containing the origin , and let @xmath380 .
set @xmath429 , @xmath430 , and define the function @xmath386 we aim at showing that @xmath431 first observe , that this inequality clearly implies the result of the theorem .
indeed , from ( [ laplace - u - t - asm ] ) and the least action principle formulated above we obtain @xmath432 on @xmath0 .
next , since @xmath418 and the origin lies in @xmath403 , we must have @xmath419 .
due to construction we also have @xmath374 , hence @xmath433 and the result follows .
it is thus left to establish ( [ laplace - u - t - asm ] ) , which we do next .
fix any @xmath65 .
there are two possibilities
. * case 1 .
* @xmath389 . similarly to case 1 of theorem [ thm - monotonicity ] we get @xmath434 where the last inequality is due to definition of @xmath10 . * case 2 .
* @xmath391 .
exactly , as in case 2 of theorem [ thm - monotonicity ] we get @xmath435 the proof of the theorem is completed .
for each integer @xmath422 let @xmath424 and @xmath436 be respectively the odometer , and final configuration of chips for initial distribution @xmath423 .
define @xmath437 and @xmath438 where @xmath439 .
it was proved by pegden and smart @xcite that there exists a non - negative and compactly supported function @xmath440 , @xmath441 satisfying @xmath442 such that @xmath443 locally uniformly in @xmath444 , @xmath445 weak * in @xmath446 and @xmath447 in the sense of distributions , where @xmath60 is continuous laplacian and @xmath88 is dirac delta at the origin . the reader is referred to @xcite for the details . concerning asm , we prove the following statement .
[ thm - asm ] for @xmath278 as above set @xmath448 and let @xmath449 the following is true : * for any @xmath450 with the property that @xmath451 is non - zero and is collinear to any of the vectors of @xmath452 , we have @xmath453 * @xmath454 is a locally lipschitz graph .
the reason for defining @xmath448 is mainly for the mathematical rigour in the proof .
apart from monotonicity of the limiting odometer function , this result shows that the boundary of the scaling limit is lipschitz .
we postpone the proof of this theorem until subsection [ subsec - scaled - od ] .
this method seems to be promising also for abelian sandile model with background heights @xcite .
the main aim of this section is to study the growth of the boundary sandpile process for point masses .
it seems intuitively clear that the value @xmath80 of boundary capacity should influence the size of the visited sites . specifically , the smaller @xmath80 the greater number of boundary points will be necessary in the stable configuration , resulting in a larger domain of visited sites .
our aim is to determine , for initial distributions concentrated at the origin , the size of the boundary capacity which will leave a chance for existence of a scaling limit . to proceed , let @xmath455 be the stabilizing pair for @xmath361 , with @xmath456 large .
let @xmath457 be the mass distribution in the stable configuration .
by definition we have that @xmath457 is concentrated on @xmath458 , and the odometer function @xmath459 satisfies @xmath460 after scaling the lattice by a small factor @xmath461 we will get @xmath462 where @xmath463 is the scaled odometer defined on @xmath464 . the scaling limit amount to taking @xmath465 .
we then want @xmath423 to converge in the sense of distributions to a continuous dirac at the origin .
since the volume element at the scale @xmath466 is @xmath467 we need @xmath468 hence the correct order of scaling would be @xmath469 .
but as @xmath470 we want @xmath471 to neither shrink to a lower dimensional object , nor escape to infinity , hence @xmath472 should be of `` uniform size '' @xmath473 .
the latter will be determined by @xmath80 .
indeed , the argument of lemma [ lem - balls - inside - out ] shows that for given @xmath80 , and @xmath456 large , @xmath472 would have size @xmath64 where @xmath474 .
but for the size @xmath64 we require @xmath475 , therefore @xmath476 is the only choice for boundary capacity to facilitate this requirement . in the light of these arguments , in dimension @xmath1
we will set @xmath477 for initial mass @xmath456 at the origin .
[ lem - mono - growth ] for @xmath478 let @xmath472 be the set of visited sites of @xmath479 .
we have the following assertions : * if @xmath480 then @xmath481 . * @xmath482 .
we start with ( a ) .
let @xmath483 for some @xmath111 . since the laplacian is a linear operator ( cf . )
, there is a collection of weights @xmath484 with all @xmath485 such that for any mass @xmath486 at the origin and @xmath487 as target set , each site @xmath488 receives mass @xmath489 .
by definition we have that @xmath490 is stable , hence @xmath491 and we get @xmath492 for all @xmath493 . from here
it follows @xmath494 which means that @xmath495 is stable for mass @xmath123 at the origin .
but as @xmath472 is the smallest among all stable domains for mass @xmath123 in view of theorem [ thm - canonical ] , we get @xmath496 , completing the proof of part ( a ) . we proceed to part ( b ) .
let @xmath478 be fixed , then in view of ( a ) we have @xmath496 for any @xmath497 .
hence from lemma [ lem - interim ] to produce @xmath490 we may topple mass @xmath498 from the origin to @xmath458 and then stabilize @xmath458 in @xmath495 . as in part ( a ) we know that each @xmath499 gets mass @xmath500 . therefore , for @xmath497 large enough , we will get @xmath501 for all @xmath114 . as a result
, all points on @xmath458 have to topple in order to stabilize @xmath495 .
this proves ( b ) , and finishes the proof of the lemma .
[ lem - no - touch ] for @xmath478 let @xmath282 be the stabilizing pair of @xmath502 .
if @xmath503 is a subset of @xmath49 containing the origin in its interior and @xmath504 is such that @xmath505 then @xmath506 .
assume for contradiction that there exists @xmath507 , and consider the function @xmath508 with @xmath509 as above .
then clearly @xmath510 in @xmath511 and @xmath512 on @xmath513 , hence by dmp we get @xmath514 in @xmath503 , i.e. @xmath515 in @xmath503 .
since both @xmath10 and @xmath213 vanish at @xmath516 we obtain @xmath517 where the last inequality follows by stability of the sandpile .
this leads to a contradiction and completes the proof .
for @xmath62 let @xmath518 be the green s function with pole at the origin for the discrete ball @xmath66 defined in ( [ def - z - r ] ) , i.e. @xmath519 [ lem - greens - laplace ] there exist constants @xmath520 depending on dimension only , and @xmath521 depending on @xmath522 and dimension @xmath523 , such that for any @xmath524 one has @xmath525 the proof is a straightforward consequence of the random walk counterpart of the problem .
let @xmath526 be the fundamental solution to @xmath60 in @xmath0 , i.e. @xmath527 for all @xmath38 .
the following asymptotics is well - known @xcite , @xcite , @xcite @xmath528 where @xmath529 is a constant , and @xmath530 is the volume of the unit ball in @xmath7 . by (
* proposition 4.6.2 ) for each @xmath531 we have @xmath532 , \ ] ] where @xmath533 is the first exit time from @xmath534 of a simple random walk @xmath535 on @xmath0 started at @xmath531 , i.e. @xmath536 . expanding the expectation in ( [ g - r ] ) and using the fact that the simple random walk started inside a finite set of @xmath0 will almost surely exit it in finite time , we get @xmath537 \big ( g(x,0 ) - g(z,0 ) \big).\ ] ] for @xmath531 satisfying @xmath538 , applying ( [ g - x - y ] ) to the last formula for @xmath518 we obtain @xmath539 where @xmath540 \begin{cases } \log \frac{|z|}{|x| } , & \text{$d=2 $ } , \\
\frac{1}{|x|^{d-2 } } - \frac{1}{|z|^{d-2 } } , & \text{$d \geq 3 $ } , \end{cases}\ ] ] and @xmath541 is a constant depending on @xmath523 .
now things are reduced to some elementary computations .
we start with a lower bound of the lemma .
consider the case of @xmath542 .
fix a point @xmath543 such that there exists @xmath544 with @xmath545 .
let @xmath546 .
in view of the definition of @xmath66 , the neighbour of @xmath547 having the largest norm among all 4 neighbours must be on the boundary of @xmath66 . since we also have @xmath548 , it follows that as @xmath549 one can take @xmath550 for @xmath551 and @xmath552 for @xmath553 .
assume we have @xmath554 , i.e. @xmath551 . then @xmath555 observe that @xmath556 since the random walk can exit from @xmath534 in one step through @xmath549 if started at @xmath547 .
on the other hand , from the definition of @xmath66 we have @xmath557 for any @xmath531 and any @xmath558 , hence all terms in the sum of @xmath559 are non - negative .
using this , for @xmath560 and @xmath549 as above we get @xmath561 this implies @xmath562 for any @xmath563 having a neighbour on @xmath564 .
since @xmath518 vanishes on @xmath564 the bound of the lemma follows .
the lower bound in case of @xmath565 is handled similarly .
the upper bound of the lemma follows directly from ( [ g1])-([g2 ] ) and definition of the ball @xmath66 .
the details , which we will omit , are elementary .
the proof of the lemma is complete .
[ lem - balls - inside - out ] there are dimension dependent constants @xmath521 such that if @xmath472 is the set of visited sites of @xmath479 , with @xmath456 large , then @xmath566 by lemma [ lem - mono - growth ] part ( b ) every site of @xmath0 is eventually visited by the sandpile if we keep increasing @xmath123 .
thus we will assume that @xmath456 is so large that the discrete ball @xmath567 is inside @xmath472 , where @xmath522 is fixed from lemma [ lem - greens - laplace ] .
let us also remark , that if for some @xmath456 we have the inclusion @xmath568 , then for all @xmath569 we get @xmath570 in view of lemma [ lem - mono - growth ] part ( a ) .
let @xmath518 be as above ( see ) .
then , the function @xmath571 satisfies @xmath572 from lemma [ lem - greens - laplace ] we have @xmath573 everywhere on @xmath564 .
observe that if @xmath574 and @xmath123 and @xmath64 are chosen so that @xmath575 on @xmath564 , then by lemma [ lem - no - touch ] we have @xmath576 , i.e. @xmath577 stays strictly inside @xmath472 .
we now choose a sequence of radii @xmath578 , where @xmath522 is as above , and for each @xmath114 we have @xmath579 and @xmath580 , where the latter simply means that we enlarge the balls by 1 lattice step at most .
given this , as long as we have @xmath581 we get @xmath582 .
we can therefore increase the radius up to @xmath583 obtaining by so the first inclusion of the current lemma .
the upper bound can be obtained similarly , by starting with a ball of sufficiently large radius , containing @xmath472 , then shrinking the ball by at most 1 lattice step at a time and using the upper bound of lemma [ lem - greens - laplace ] instead .
the proof of the lemma is complete .
our next result gives a uniform lower bound on the odometer function , which will be needed in the analysis of the shape of scaling limit .
for the proof we need a discrete version of harnack s inequality .
[ thm - harnack ] for any @xmath584 there exists a constant @xmath585 such that if @xmath586 is harmonic in @xmath534 then @xmath587 for any @xmath588 . [ lem - bounds - below ]
for any @xmath589 small there exists a constant @xmath590 depending on @xmath591 and dimension @xmath523 such that for any @xmath456 and each @xmath592 satisfying @xmath593 one has @xmath594 where @xmath455 is the stabilizing pair of @xmath479 .
we first observe that the bound of the lemma holds near the origin , i.e. there exist dimension dependent constants @xmath595 and @xmath596 such that for any @xmath456 one has @xmath597 indeed , by lemma [ lem - balls - inside - out ] and dmp we have @xmath598 for some constant @xmath599 , where the green s function @xmath48 is defined in ( [ green - def ] ) .
thanks to this bound , follows directly from asymptotics of green s function proved in ( * ? ? ?
* proposition 1.5.9 ) for @xmath565 , and ( * ? ? ? * proposition 1.6.7 ) for @xmath542 .
next , we reduce the general case of the lemma to by constructing a harnack chain leading from a given point to the ball @xmath600
. let @xmath601 and assume @xmath602 , since otherwise the estimate follows by .
suppose @xmath603 , and denote @xmath604 , where @xmath605 is fixed from ( [ u - n - large - near-0 ] ) and @xmath591 from the formulation of the lemma .
we get that the ball @xmath606 lies inside @xmath607 . by @xmath608 denote the orthogonal projection of @xmath609 onto @xmath610 , i.e. @xmath611 by directional monotonicity of theorem [ thm - monotonicity ] applied to the direction @xmath612 , and axial symmetry of the odometer given by corollary [ cor - o - symm ] , we have @xmath613 \big ) \cap { { \mathbb z}}^d \subset v_n.\ ] ] if the cylinder @xmath614 intersects the ball @xmath615 , the estimate of the lemma follows by harnack s inequality of theorem [ thm - harnack ] and upper estimate of lemma [ lem - balls - inside - out ] , where the latter is used to estimate the length of harnack chain . otherwise , if @xmath616 , denote @xmath617 , and @xmath618 . by construction @xmath619 .
we have , due to harnack s inequality , that @xmath620 with a constant @xmath621 , hence it is enough to prove the lemma for @xmath622 replaced by @xmath623 .
for that , we apply the same argument as we had for @xmath622 to @xmath623 .
since @xmath623 has at most @xmath624 non - zero coordinates , this reduction procedure on coordinates will terminate in at most @xmath624 number of steps , where in the last step , the corresponding cylinder ( [ cylinder ] ) will intersect the ball @xmath625 implying the desired estimate of the lemma .
the proof is complete .
the purpose of this section is the study of scaling limit of the model when the initial distribution is concentrated at a point .
we will show that after proper scaling of mass and the underlying lattice , there is a convergence along subsequences .
the approach is via uniform gradient bounds for the odometers .
combined with some results of the previous section , we will prove a certain non - degeneracy result of the converging shapes , as well as lipschitz regularity of the boundary of scaling limit . throughout this section
, we assume that initial distribution of the sandpile equals @xmath423 , where @xmath456 and boundary capacity of the model is @xmath626 . for all @xmath456 , by @xmath424
we denote the corresponding odometer function and by @xmath472 the set of visited sites .
let @xmath631 be the green s function for @xmath472 , i.e. @xmath632 for any @xmath633 and @xmath634 if @xmath635 or @xmath636 . from (
* proposition 4.6.2 ) for @xmath633 we have @xmath637 - g(x , y),\ ] ] where as before @xmath23 is defined from ( [ g - x - y ] ) , and @xmath638 is the first exit time from @xmath639 of the simple random walk @xmath535 started at @xmath58 . from the definition of @xmath640 we have @xmath641 where we have used the asymptotics ( [ g - x - y ] ) and mean - value theorem along with the choice of @xmath58 and @xmath57 to bound the first summand . from lemma [ lem - balls - inside - out ] and the choice of @xmath642
we get that @xmath643 for any @xmath644 .
this , coupled with ( [ g - x - y ] ) and mean - value theorem , for any @xmath644 implies @xmath645 combining this estimate with ( [ a1 ] ) we obtain @xmath646 we get from the definition of @xmath424 that @xmath647 for all @xmath648 , which together with ( [ a2 ] ) completes the proof of the lemma .
recall that @xmath651 is the set of visited sites for mass @xmath456 .
define @xmath652 where @xmath653 is the combinatorial distance . for
each @xmath44 consider the discrete derivative of @xmath10 , namely @xmath654 with @xmath65 .
clearly @xmath655 is harmonic in @xmath656 .
now , in view of the stability of the sandpile , for any @xmath648 such that @xmath657 we have @xmath658 with a constant @xmath659 . on the other hand , by lemma [ lem - grad - outside-0 ] we have @xmath660 if @xmath661 .
since @xmath662 in the interior of @xmath656 , by dmp we get @xmath663 in @xmath656 .
the same bound obviously works for @xmath664 .
we have thus proved 1-step lipschitz bound , i.e. the estimate of the proposition if @xmath58 and @xmath57 are lattice neighbours . for the general case , take any @xmath642 such that @xmath649 , and consider the shortest lattice path connecting @xmath58 and @xmath57 and staying outside the ball @xmath665 .
namely , let @xmath666 where @xmath667 and @xmath668
. clearly such path exists .
it is also clear that for the length of the path we have @xmath669 where equivalence is with dimension dependent constants . using this and
the 1-step lipschitz bound already proved above , we obtain @xmath670 completing the proof of the proposition . for @xmath478 set @xmath671 , and define the scaled odometer by @xmath672 where @xmath673 . let also @xmath674 be the scaled set of visited sites for mass @xmath123 .
clearly , @xmath675 is supported in the interior of @xmath676 . moreover , in view of lemma [ lem - balls - inside - out ] we have that the sets @xmath677 , are uniformly bounded and contain a ball of some fixed radius . in order to study the scaling limit of the model , we need to extend each @xmath683 to a function defined on @xmath7 .
we will use a standard extension of @xmath683 which preserves its @xmath52-laplacian .
namely , for fixed @xmath684 define a function @xmath685 , where for each @xmath686 and any @xmath687 we have set @xmath688 . clearly @xmath689 for all @xmath687 .
note that we are using the same notation for extended odometers . in what follows @xmath683 will stand for this extension . also , for a given set @xmath690 we write @xmath691 for the interior of @xmath692 in the next theorem .
the following is our main result concerning scaling limit of the model . * @xmath695 locally uniformly in @xmath696 , * @xmath697 in @xmath698 in the sense of distributions , where @xmath88 is the dirac delta at the origin and @xmath60 denotes the continuous laplace operator , * if @xmath699 is the support of @xmath278 then , @xmath700 * setting @xmath448 , and considering the set of vectors @xmath701 for any @xmath702 with the property that @xmath703 is non - zero and is collinear to any of the vectors of @xmath452 , we have @xmath704 * the boundary of @xmath705 is locally a lipschitz graph . fix
@xmath706 small , and for @xmath678 set @xmath707 extend each @xmath683 as 0 outside @xmath676 . from proposition [ lem - lipschitz - est ] and definition of @xmath683 , there exists a constant @xmath708 independent of @xmath16 such that @xmath709 we have @xmath710 , hence for any @xmath711 thanks to ( [ u - h - lipschitz ] ) we obtain @xmath712 where the second inequality with a constant @xmath659 follows from lemma [ lem - balls - inside - out ] .
thus each @xmath683 , @xmath684 , when restricted to @xmath713 , is bounded uniformly in @xmath16 due to and is @xmath714-lipschitz in view of .
next , we extend each @xmath683 from @xmath713 to a lipschitz function on @xmath7 having the same lipschitz constant .
namely , for a given @xmath466 define @xmath715 this is a well - known method of lipschitz extension , and it is not hard to verify that @xmath716 is a @xmath714-lipschitz function on @xmath7 ( see e.g. ( * ? ? ?
* theorem 2.3 ) ) and coincides with @xmath683 on @xmath713 .
moreover , by construction we have that the family @xmath717 is uniformly bounded and is non - negative everywhere . observe also that for @xmath684 , @xmath718 and @xmath719 ( see ) by construction we have latexmath:[\[\label{u - h - u - h - est } by arzel - ascoli there is sequence @xmath693 as @xmath101 , and @xmath714-lipschitz function @xmath721 such that @xmath722 locally uniformly in @xmath7 .
due to ( [ u - h - u - h - est ] ) we get @xmath723 locally uniformly in @xmath724 . obviously @xmath725 on @xmath7 . for the support of @xmath726 we have @xmath727 to prove this take any @xmath58 from the _ l.h.s .
_ of ( [ supp - outside-2delta ] ) .
since @xmath726 is continuous on @xmath7 there is @xmath728 such that the closed ball @xmath729 and @xmath730 on @xmath731 for some @xmath732 . by construction
we have that @xmath733 converges to @xmath726 uniformly on @xmath731 , hence there exists @xmath734 large enough such that @xmath735 on @xmath731 for any integer @xmath736 .
but since @xmath733 agrees with @xmath737 on @xmath738 we get @xmath739 for any @xmath740 .
this implies @xmath741 on @xmath742 hence @xmath743 for all @xmath736 . in particular
, @xmath58 is from the _ r.h.s .
_ of ( [ supp - outside-2delta ] ) . to see the reverse inclusion fix any @xmath58 from the _ r.h.s .
_ of ( [ supp - outside-2delta ] ) .
there exists @xmath744 large and @xmath728 small such that @xmath745 .
we get , in particular , that @xmath746 for all @xmath736 . hence , by lemma [ lem - bounds - below ] and definition of @xmath737 one gets @xmath747 on @xmath748 with a constant @xmath596 uniform in @xmath736 .
this bound , together with ( [ u - h - u - h - est ] ) implies @xmath749 , accordingly @xmath58 lies in the _ l.h.s .
_ of ( [ supp - outside-2delta ] ) .
this completes the proof of ( [ supp - outside-2delta ] ) .
we next prove that @xmath726 is harmonic on @xmath750 .
fix any @xmath751 such that @xmath752 .
since @xmath726 is continuous , there exists a closed ball @xmath753 .
let @xmath754 be any . since @xmath726 is continuous , to prove its harmonicity in @xmath755 it suffices , due to weyl s lemma , to show that @xmath756 where @xmath60 is the usual ( continuous ) laplace operator in @xmath7 .
the latter follows by @xmath757 where the first equality comes from definition of @xmath726 and smoothness of @xmath758 , the second is in view of ( [ u - h - u - h - est ] ) , the third one follows from discrete integration by parts .
the last equality is a consequence of the fact that for large enough @xmath744 due to one has @xmath759 for all @xmath736 , which , in particular , implies that @xmath760 on @xmath761 for all @xmath736 . *
@xmath763 locally uniformly outside any neighbourhood of 0 , * @xmath764 , * @xmath765 in @xmath766 , * @xmath278 is lipschitz continuous outside any neighbourhood of the origin , where lipschitz norm depends on the neighbourhood .
these properties give us parts ( i ) and ( iii ) of the theorem , as well as the claim of ( ii ) except at the origin .
to finish the proof of part ( ii ) it remains to study the laplacian of @xmath278 at 0 . to this end ,
let @xmath767 be the fundamental solution for @xmath52 in @xmath464 , i.e. @xmath768 . here
as well , following the convention above , suppose that @xmath767 is extended to @xmath7 so that to preserve its @xmath52-laplacian .
it is well - known ( see @xcite ) that @xmath769 as @xmath470 locally uniformly in @xmath696 , where @xmath770 is the fundamental solution to continuous laplacian .
observe , that in view of ( b ) and lemma [ lem - balls - inside - out ] there is @xmath728 small such that the ball @xmath771 .
using this and the definition of @xmath737 we have @xmath772 but since both @xmath737 and @xmath773 converge locally uniformly away from 0 to continuous functions , then the difference @xmath774 on @xmath775 for all @xmath111 where @xmath776 is some large constant .
by dmp we get that @xmath774 in @xmath777 , hence , the limit @xmath778 is also bounded by @xmath37 in @xmath779 . but as @xmath780 in @xmath779 we get that @xmath781 is a removable singularity for @xmath782 , in particular @xmath783 at 0 , and the proof of part ( ii ) is now complete .
it remains to show ( v ) , the last assertion of the theorem .
we will show that at each @xmath784 there exists a double cone , with its size ( opening and height ) independent of @xmath622 , having vertex at @xmath622 and intersecting @xmath785 at @xmath622 only is the 0-level set of @xmath278 which is a lipschitz function on compact subsets away from the origin , and in particular in a neighbourhood of @xmath324 .
but this fact alone is not enough to conclude that @xmath324 is lipschitz , see for instance @xcite . ] .
this clearly implies ( v ) . applying the monotonicity of ( iv ) in directions
@xmath786 we obtain @xmath787 for all @xmath42 . due to this symmetry , to prove ( v ) it is enough to treat the part of @xmath324 where all coordinates are non - negative .
in addition , we will also assume that the point @xmath788 satisfies @xmath789 for all @xmath790 .
the rest of the cases are similar .
observe that due to ( iii ) and lemma [ lem - balls - inside - out ] there is a constant @xmath590 such that @xmath791 now the choice of @xmath622 and ( [ ball - inside ] ) together imply @xmath792 we first determine a certain monotonicity region for @xmath622 .
let @xmath793 be a finite set of vectors containing @xmath612 and having the rest of its elements chosen by the following rule : for each @xmath794 it is easy to check from definition of @xmath267 that it contains @xmath523 linearly independent vectors .
for a given @xmath800 consider the halfspace @xmath801 .
we have the following monotonicity of @xmath278 in @xmath802 .
indeed , decomposing @xmath804 into tangential and normal components in @xmath802 , one gets @xmath805 , where @xmath164 , @xmath806 and @xmath807 is the same for both @xmath808 due to assumption .
then , @xmath809 , hence @xmath367 if and only if @xmath810
but since @xmath811 the latter reduces to @xmath812 .
it remains to apply the monotonicity result of part ( iv ) to get .
we thus obtain that @xmath278 is non - increasing in @xmath802 in the normal direction . as a direct corollary of ( [ mono - plane ] ) the cone
@xmath816 inherits the following property .
if @xmath817 and @xmath818 where @xmath819 , @xmath820 for all @xmath493 , and @xmath821 is the cardinality of @xmath267 , then @xmath822 along with @xmath823 consider as well the cone generated by @xmath267 @xmath824 since @xmath267 has @xmath523 linearly independent vectors , the cone @xmath825 is @xmath523-dimensional .
we claim that @xmath826 is the sought cone , for which we need to show that the truncated cone @xmath816 has uniform size and that @xmath278 is positive in the interior of @xmath816 and is zero in the interior of @xmath827
. we will only show the former claim , as the latter follows similarly .
it is clear , thanks to ( [ x-0-is - inside ] ) , that @xmath816 is of uniform size .
now assume for contradiction , that there is @xmath828 such that @xmath829 .
consider the cone @xmath830 .
let us see that @xmath831 the first assertion of ( [ 2claims ] ) follows directly from since @xmath832 by definition and any element @xmath833 is from @xmath823 as well , and has the form @xmath834 with all @xmath819 .
the second one is simply a consequence of the definition of @xmath372 .
armed with ( [ 2claims ] ) the proof of ( v ) follows readily , since we simply get that @xmath278 is zero in an open neighbourhood of @xmath622 , which violates the condition that @xmath835 .
this contradiction finishes the proof of ( v ) , and the proof of the theorem is now complete .
assertion ( a ) follows from convergence result of pegden - smart discussed in subsection [ subsec - asm ] in conjunction with theorem [ thm - monotonicity - asm ] . the proof of part ( b ) follows from part ( a ) by the same argument as for part ( v ) of theorem [ thm - scaling - limit ] .
the proof is complete .
it is clear from the proof of theorem [ thm - scaling - limit ] that for any sequence of scalings @xmath836 , where @xmath837 , one may extract a subsequence @xmath838 such that the corresponding sequence of scaled odometers will have a limit , call it @xmath839 , satisfying the same properties as the function @xmath278 of theorem [ thm - scaling - limit ] .
in particular , by theorem [ thm - scaling - limit ] ( ii ) we have that any such @xmath839 is @xmath840 on the set where it is positive , and is lipschitz continuous up to the boundary of that set . however , we do not know if scaling limits generated by different scaling sequences must coincide . for several lattice - growth models
the existence of scaling limit , and in some cases its geometry , are known ; for divisible sandpile see @xcite , @xcite and @xcite , for rotor - router see @xcite , @xcite , @xcite , and @xcite , and for abelian sandpile see @xcite . the internal diffusion limited aggregation ( idla ) , which is a stochastic growth model ,
is studied in @xcite , @xcite , @xcite , @xcite . for the idla in critical regime
, one may consult @xcite .
growth models involving continuous amounts of mass similar to divisible sandpile model , are studied in @xcite , @xcite , and @xcite , where the last paper considers a non - abelian growth model .
it is interesting to observe , that the strong flat patterns appearing on the boundary of a sandpile as can be seen in figure [ fig-1 ] , might be a result of a significant increase of the role of the free boundary , rather than the pde problem in the interior .
it is remarkable that a similar geometry of the boundary as in figure [ fig-1 ] , in particular the flatness , arises also from the classical abelian sandpile , see figure [ fig-2 ] , if one treats the boundary of the growth cluster likewise , i.e. allowing each boundary point to accumulate many particles however not exceeding the given threshold .
nevertheless , the pde problem of the abelian sandpile in the set of visited sites , is very different from that of the boundary sandpile considered here .
it seems very interesting to understand the emergence of the flatness on the boundary rigorously .
yet another surprising property of the boundary sandpile is the discontinuity of its geometry under small perturbations of the initial mass distribution , and a certain tendency of the sandpile dynamics towards generating convex sets ( see figure [ fig-3 ] ) .
a rigorous explanation of these phenomena seems to be a very interesting problem .
* problem 3 .
* is there a boundary sandpile type process , leading to that the total mass of the system is being redistributed onto the combinatorial free boundary ( possibly with a different background rules ) which has a sphere as its scaling limit ?
in fact , this problem does not exclude , as a possible candidate , the boundary sandpile process considered in this paper .
it is apparent from numerical simulations ( see figure [ fig-1 ] ) , that the shape of @xmath94 divaricates from a sphere . however , we do not have a rigorous proof of this fact .
gustafsson , b. : direct and inverse balayage - some new developments in classical potential theory .
proceedings of the second world congress of nonlinear analysts , part 5 ( athens , 1996 ) nonlinear anal .
30 ( 1997 ) , no . 5 , 2557 - 2565 | we introduce a new lattice growth model , which we call boundary sandpile . the model amounts to potential - theoretic redistribution of a given initial mass on @xmath0 ( @xmath1 ) onto the boundary of an ( a priori ) unknown domain .
the latter evolves through sandpile dynamics , and has the property that the mass on the boundary is forced to stay below a prescribed threshold .
since finding the domain is part of the problem , the redistribution process is a discrete model of a free boundary problem , whose continuum limit is yet to be understood .
we prove general results concerning our model . these include canonical representation of the model in terms of the smallest super - solution among a certain class of functions , uniform lipschitz regularity of the scaled odometer function , and hence the convergence of a subsequence of the odometer and the visited sites , discrete symmetry properties , as well as directional monotonicity of the odometer function .
the latter ( in part ) implies the lipschitz regularity of the free boundary of the sandpile . as a direct application of some of the methods developed in this paper , combined with earlier results on classical abelian sandpile
, we show that the boundary of the scaling limit of abelian sandpile is locally a lipschitz graph . |
in previous papers @xcite , we reported on the spectra of four comets observed by the _ far ultraviolet spectroscopic explorer _ ( _ fuse _ ) . launched in june 1999 , _ fuse _ operated through october 2007 providing an orbiting capability with a spectral resolution better than 0.4 in the wavelength range 9051187 together with very high sensitivity to weak emissions making possible both the search for minor coma species and extensive temperature and density diagnostics for the dominant species . our initial reports on the three comets observed in 2001 focused on three hopfield - birge band systems of co not previously observed in comets , the @xmath6 @xmath1(0,0 ) band at 1150.5 , the @xmath0 @xmath1(0,0 ) band at 1087.9 , and the weak @xmath7 @xmath1(0,0 ) band at 1076.1 , three lines of h@xmath8 fluorescently pumped by solar lyman-@xmath9 radiation , and upper limits to emission from and .
the rotational envelopes of the co bands are resolved and appear to consist of both cold and warm components , the cold component , with a rotational temperature in the range 5575 k , accounting for over 75% of the flux and presumably due to native co released by the nucleus .
the warm ( @xmath2500 k ) component suggested a photodissociation source but this was considered unlikely because the low rate for this process in the solar radiation field would have required a very high relative abundance of .
emission at 1031.9 and 1037.6 was searched for but only the 1031.9 line was marginally detected in c/2000 wm1 at about the level predicted by a model used to explain the x - ray emission from comets as being produced by charge exchange between solar wind ions and cometary neutrals @xcite .
the higher signal / noise ratio in the spectrum of comet c/2001 q4 , observed in april 2004 , enabled an accurate determination of the wavelength of this feature which suggested that it was the ( 1,1 ) q(3 ) werner ( @xmath10 ) line , which can be fluorescently pumped by either solar @xmath4989.8 or @xmath41031.9 .
the latter requires the to be in the first vibrationally excited state @xcite .
the q(3 ) lines of the ( 1,3 ) band at 1119.1 and the ( 1,4 ) band at 1163.8 are also detected as would be expected for fluorescent excitation .
however , the ground state abundance in the coma , estimated from the solar lyman-@xmath9 pumped fluorescence of the lyman ( @xmath11 ) ( 6,1 ) p1 line , is insufficient for excitation by solar ( a factor of 10 weaker than ) , which implies the formation of in the @xmath12 state and excitation by solar . the molecular channel of the photodissociation of formaldehyde ( i.e. , the channel leading to and co products ) has been well studied both experimentally and theoretically .
its products are characterized by a non - thermal rotationally `` hot '' co molecule and a vibrationally excited molecule @xcite .
the co rotational distribution can be represented by a gaussian centered near @xmath13 with a half - width at half maximum in @xmath3 of about 20 .
fluorescence from such levels in the p - branch ( the r - branch is shortward of the detector wavelength cut - off ) is clearly seen in two of the comets observed by _ fuse _ and we propose that this feature , together with the vibrationally excited , constitute an unambiguous signature of in a cometary coma .
formaldehyde has been observed in these comets in both the infrared and millimeter spectral regions @xcite .
the capability of _ fuse _ to identify in a cometary coma provides a new means to determine its production rate relative to that of water and to compare these results with ground - based observations in the infrared and millimeter regions of the spectrum . since the thermal `` hot '' component of the observed co emission is not produced by dissociation , we return to as a possible parent , particularly as the rotational temperature is what would be expected based solely on angular momentum conservation arguments @xcite .
we suggest that within the _ fuse _ field - of - view , photoelectron impact dissociation of has a comparable rate to photodissociation and we base this conclusion on estimates of the mean photoelectron flux derived from the other prominent co band observed in these spectra , the @xmath14 ( 0,0 ) band at 1151 .
we thus have unique diagnostics in which the shapes of these molecular bands can be used to infer the column abundance of co from both its native source as well as from two of its direct molecular parents , together with the photoelectron contributions to excitation and dissociation in the inner coma .
three long period comets were observed during 2001 , c/1999 t1 ( mcnaught - hartley ) , c/2001 a2 ( linear ) , and c/2000 wm1 ( linear ) , with c/2001 a2 being the most active of the three and showing the brightest far - ultraviolet emissions . in december 2001 , two of the reaction wheels on the spacecraft failed resulting in a shutdown of observatory operations until a new pointing control program could be devised and implemented a few months later .
however , no moving target observations were attempted until early in 2004 when comet c/2001 q4 ( neat ) was observed .
the observation parameters for all four comets are summarized in table [ tab1 ] .
observations of each comet were made over contiguous spacecraft orbits during which the center of the @xmath15 spectrograph aperture was placed at the ephemeris position of the comet and tracked at the predicted sky rate .
due to the extended , though not uniform , nature of the cometary emissions within the aperture , the effective spectral resolution is @xmath20.25 .
for each comet , all of the orbits of data were co - added and the extracted fluxes were converted to average brightness ( in rayleighs ) in the @xmath15 aperture .
in addition , the data were separately co - added for only the time during which the spacecraft was in earth shadow
. these `` night - only '' spectra make it possible to differentiate cometary emissions of , , , and , in second order , , from the same emissions produced in the daytime terrestrial atmosphere @xcite . for the co and emissions
this is not a problem and the entire data set for each comet was used .
all of the data were reprocessed using the final version of the _ fuse _ pipeline , _ calfuse 3.2 _ @xcite . among the enhancements incorporated in this version
were temperature - dependent wavelength corrections , time - dependent corrections to the detector effective areas , and a more complete evaluation of detector background .
the latter was particularly significant in allowing reliable extractions of the fluxes of weak emission features .
the overall absolute flux calibration , based on standard stars and adjusted for the date of observation , is better than 5% .
table [ tab1 ] also lists the extracted brightnesses of the co emission features discussed below .
the co @xmath16 ( 0,0 ) band , from the total exposure , is shown for each comet in figure [ cx1 ] .
the p and r branches appear well separated , characteristic of a rotational temperature of 5575 k. there also appear broad wings to the band indicative of a rotational temperature in the range of 500 to 600 k. the figures also show a model fit , assuming solar resonance fluorescence , obtained using the procedure described by @xcite .
the derived column densities and rotational temperatures are given in table [ tabn ] .
in the two most active comets , c/2001 a2 and c/2001 q4 , there is also an additional broad feature centered at @xmath21089.4 , that we identify as p - branch lines from high - lying @xmath3 states .
the position of the p(40 ) line , calculated using the rotational constants of @xcite , is also shown in fig . [ cx1 ] .
we note that the @xcite wavelengths have been verified experimentally only up to @xmath17 . the corresponding high @xmath3 r - lines are off the short wavelength edge of the lif2a detector .
a `` hot '' thermal rotational distribution arises in molecular photodissociation in order to conserve angular momentum when the product molecule has a much smaller moment - of - inertia than the parent molecule as is the case for @xcite .
the non - thermal component is identified with co as a dissociation product of , and the column abundance of this component , also given in table [ tabn ] , is derived assuming an equal contribution from the unobserved r - branch . in this case ,
an optically thin band fluorescence efficiency is used as the co population is spread over many rotational levels and the absorption is not likely to be optically thick . as noted above ,
the photodissociation of formaldehyde has been intensively studied , both experimentally and theoretically . the lowest dissociation threshold ( @xmath23600 ) leads to + co ( the `` molecular '' channel ) while @xmath22700 @xmath18 higher ( @xmath23300 ) , h + hco ( the `` radical '' channel )
can also be produced @xcite .
the nature of the and co products differs markedly above and below this second energy threshold .
consider first the region below the radical channel threshold .
@xcite and @xcite have used laser - induced fluorescence spectroscopy and coherent anti - stokes raman scattering to experimentally study the product state populations of co and .
@xcite found a nonthermal highly excited co rotational population with a gaussian distribution centered at @xmath19 and a half - width at half maximum of 20 @xmath3 units , but with little vibrational excitation .
later work ( e.g. , * ? ? ?
* ) showed that the shape was dependent on the energy of the absorbed photon .
it should be noted that the absorption of in this region is in discrete bands @xcite . for
, @xcite found that the primary products are the ortho ( @xmath20 and 5 ) states of @xmath12 .
for photolysis at 3390 they found the fractional population of the @xmath12 , @xmath20 level to be 11% .
@xcite extended this work to additional wavelengths and found a peak population in @xmath21 .
above the radical channel threshold , and co are found to have markedly different populations with the non - translational energy going principally to vibrational excitation of while the co rotational state population shifts to lower @xmath3 states @xcite .
@xcite demonstrated that the population of vibrationally excited , up to @xmath22 or 9 , increased as the energy of the exciting photon increased with a corresponding increase in the low-@xmath3 state population of co such that the non - translational energy was conserved .
however , the peak at @xmath19 remained and the low-@xmath3 population was still highly non - thermal .
@xcite showed that the excited vibrational levels of also had peak rotational populations at @xmath23 or 7 , depending on the photolysis energy . in the solar radiation field in the coma of a comet
, the and co state populations derived from dissociation will reflect a superposition of populations , weighted by the solar flux and absorption cross section @xcite .
such a calculation is beyond the scope of the present paper .
the broad , weak feature seen longward of the @xmath16 ( 0,0 ) band in comets c/2001 a2 and c/2001 q4 in fig . [ cx1 ] is centered near the position of the p(40 ) line of this band , and the non - thermal high-@xmath3 population of this feature is an unambiguous fingerprint of co produced from photodissociation . although the spectrum of c/2000 wm1 is significantly noisier , the data also suggest enhanced emission on the red wing of the band in this comet .
further evidence for this source is the simultaneous detection of fluorescence from vibrationally excited in this comet @xcite as well as in comets c/2001 a2 and c/2001 q4 . in figure [ h2 ] the region of the doublet in comet
c/2001 q4 is shown , and the ( 1,1 ) q(3 ) line is identified , separated by 0.06 from the position of the line at 1031.92 . the uncertainty in the _ fuse _ wavelength scale is @xmath20.01 .
the same figure also shows the corresponding ( 1,3 ) q(3 ) and ( 1,4 ) q(3 ) lines at 1119.09 and 1163.79 , respectively @xcite .
note that there is no emission at the expected position of the line at 1037.62 .
these lines are quite weak and with the exception of a 3-@xmath24 detection in comet c/2000 wm1 @xcite do not appear in the spectra of the two other comets observed by _
the absence of these lines in the other two comets is the result of the heliocentric doppler shift of the exciting solar line , the swings effect @xcite , rather than the lack of in the coma .
this can be illustrated with the solar @xmath41031.9 line profile .
@xcite have used the recent solar spectra taken by the sumer instrument on _ soho _ with an instrumental resolution of @xmath20.1 , to determine the width of this line under quiet sun conditions to be 0.202 .
this result is from an observation of a small part of the solar disk near the center , while for the calculation of cometary excitation rates the whole disk flux is more useful .
fortuitously , we are able to take advantage of a minor operational problem of _ fuse _
@xcite , a sensitivity at certain orientations on the sky of the silicon carbide ( sic ) channels to scattered solar radiation , to determine the whole disk line shape .
the observation of comet c/2001 q4 resulted in such an orientation and the sic1b and sic1a channels ( 905992 and 10051090 , respectively ) measured a large scattered solar flux in all three apertures . both the high ( hirs , @xmath25 ) and medium resolution ( mdrs , @xmath26 ) apertures obtained good s / n spectra of the lines at instrumental resolutions of 0.035 and 0.075 , respectively . from both of these channels we derived a width of 0.200 for the solar @xmath41031.9 line , in excellent agreement with the result of @xcite .
the mdrs profile is shown in figure [ solar ] , together with the doppler shifted positions of the ( 1,1 ) q(3 ) line for each comet .
note that the profile shown in the figure suggests that the solar line is not adequately represented by a single gaussian . clearly , the detection of the ( 1,@xmath27 ) lines in comet c/2001
q4 was favored by its negative heliocentric velocity and by the relatively high activity level of this comet at the time of observation .
@xcite noted that solar lyman-@xmath5 ( 1216.57 ) could pump vibrationally excited in the @xmath28 level via the lyman band ( 1,2 ) p(5 ) and r(6 ) lines , and used this mechanism to identify fluorescence from the upper level in _ hubble space telescope _ spectra of jupiter following the impact of comet shoemaker - levy 9 in july 1994 .
the same mechanism has recently been found to operate in planetary nebulae where the _ fuse _ spectra show the ( 1,1 ) p(5 ) and r(6 ) lines at 1161.82 and 1161.95 , respectively @xcite .
one of the previously unidentified lines reported in _ fuse _
spectra of comets at 1161.83 was identified by @xcite as a blend of these two lines ( see fig . [ h2 ] ) .
the vibrationally excited would most likely come from rather than photodissocation as proposed by @xcite this feature has roughly the same brightness in comets c/2001 a2 and c/2001 q4 , and we note that the brightnesses of the co p(40 ) feature in these two comets are also comparable . from the branching ratios given by @xcite , we would also expect to see the ( 1,1 ) r(3 ) line at 1148.70 and the ( 1,1 ) p(8 ) line at 1183.31 from the same excited @xmath3 levels , and indeed we do .
the latter is considerably weaker than the former due to the lower population of the even-@xmath3 ( para ) states as expected .
@xcite also identified several other cometary lines as being pumped by solar lyman-@xmath5 from higher vibrational and rotational levels of .
almost all of the strong unidentified features tabulated by @xcite are amongst them .
our analysis supports the conclusion of @xcite that the vast majority of emission features in the _ fuse _ spectra of comets can be attributed to .
however , unlike @xcite , who proposed that the excited vibrational and rotational levels of arise from the photodissociation of , we suggest instead that these excited states are produced during the photodissociation of because their presence in the cometary coma is consistent with the level populations measured during photodissociation in the laboratory by @xcite and @xcite . the co @xmath29 ( 0,0 ) band at 1151 is weaker than the @xmath16 ( 0,0 ) band by a factor of 3 to 4 and thus recorded with poorer signal / noise in all four comets as shown in figure [ bx1 ] .
it is surprising that this band is observed at all since its fluorescence efficiency is 44 times lower than that of the @xmath16 ( 0,0 ) band ( the predicted spectra using the co column densities given in table [ tabn ] are also shown in the figure ) .
there must be an additional source of excitation , and since the band shape is consistent with the low temperature co component , that can only be electron impact excitation of co. photoelectrons in the inner comae of comets have been considered by several authors ( e.g. , * ? ? ?
* ; * ? ? ?
* ) , and were detected in the _ vega _ fly - by of 1p / halley @xcite , but their role in excitation , dissociation and ionization is difficult to quantify .
an equivalent to the fluorescence efficiency , or g - factor , may be defined by an electron excitation rate per molecule per second , @xmath30 : @xmath31 where @xmath32 is the photoelectron flux in electrons s@xmath33@xmath34sr@xmath33ev@xmath33 and @xmath35 is the cross section in @xmath36 . in general , the photoelectron flux is a function of cometocentric distance and is anisotropic so that @xmath30 will vary over the spectrograph field - of - view . rather than attempt to use modeled photoelectron fluxes ,
we instead use the observed co @xmath29 ( 0,0 ) band brightnesses to derive an effective @xmath30 over the _ fuse _ aperture but then use the modeled electron energy distribution to determine the relative excitation rates for other processes . in particular , we are interested in the electron impact dissociation rate of .
relevant cross sections in analytic form for electron impact on co and are available from the compilation of @xcite .
figure [ cross ] shows the cross sections for electron impact excitation on co of the two bands under discussion and that for electron impact dissociation of into the metastable @xmath37 state of co. a typical photoelectron energy dependence , from @xcite , is also shown .
the @xmath37state is the upper level of the co cameron bands , whose excitation in comets has been discussed by @xcite . the total dissociation cross section has been studied by @xcite .
these authors note that the branching ratio into the @xmath37 state of co is 27% of the total electron dissociation rate , the same as for solar photodissociation @xcite .
the derived values of @xmath30 , reduced to 1 au , are given in table [ tabg ] . for comparison ,
the optically thin solar fluorescence efficiencies are also given .
figure [ cross ] shows that the steep increase of photoelectron flux towards lower energies favors the @xmath29 transitions over the @xmath16 .
even so , the electron impact contribution to @xmath38 state excitation will reduce the derived co column abundances by 515% and this is reflected in the values given in table [ tabn ] .
these numbers remain uncertain as both the photoelectron energy distribution and flux will vary with increasing column density towards the nucleus
. this will also affect the ratio of total dissociation to that fraction leading to the @xmath37 state of co. the @xmath30 values for these processes , for the photoelectron model adopted , are also given in table [ tabg ] .
for comparison with photodissociation at 1 au , the solar maximum value given by @xcite is @xmath39 molecule@xmath33 s@xmath33 , so electron impact dissociation and photodissociation are comparable .
evidence for a significant role of photoelectrons in the inner coma has often been noted in the literature .
two particular cases in which photodissociation and photoionization were inadequate to account for the abundance of daughter products are the production of atomic carbon in the metastable @xmath40d state from co @xcite and the high concentration of close to the nucleus seen in early _ iue _ spectra @xcite .
the _ fuse _ spectra provide additional evidence in the strong @xmath40d
@xmath40d@xmath41 line at 1152.16 , seen in fig .
[ bx1 ] , which also can not be accounted for by solar resonance scattering alone . the lower state of this transition , the metastable @xmath40d state , is an abundant product of photodissociation of , oh , and co , and is the upper level of the oxygen `` red '' lines at 6300 and 6364 , seen in many comets and often used as a surrogate for the water production rate @xcite .
the column density of the non - thermal component of co was modeled assuming an parent using the vectorial model of @xcite with solar photodestruction rates given by @xcite .
@xcite have also evaluated the photodissociation lifetimes ( has two other dissociation channels , h + hco , and h + h + co ) and find the rate into + co to be 25% lower than that of @xcite while the total rate is 7% lower .
@xcite find a total photodissociation lifetime in close agreement with that of @xcite the model assumes a nucleus source for the .
even if a distributed source were present ( see section [ dist ] below ) , the size of the _ fuse _ aperture translates to a few thousand km projected on the sky and such a source can be ignored for our purposes . to determine production rates relative to water
, we use values of @xmath42 taken from the recent literature ; values derived from soho / swan measurements of cometary lyman-@xmath5 emission by @xcite , and those derived from odin submillimeter observations of directly by @xcite .
where they overlap , these two sets of values are in agreement to @xmath22025% .
we adopt the values from @xcite because of their finer sampling grid , allowing us to use data from the same day as the _ fuse _ observations .
we use the deconvolved model results from @xcite rather than the daily averages given . for c/2001
q4 we use a value derived from near - simultaneous _ hst _ observations of oh emission , which is close to that reported by @xcite for c/2001 a2 we note a significant difference from the value of @xmath42 cited in the initial reports of the _ fuse _ observations by @xcite and @xcite .
the earlier estimate was based on a visual outburst two days prior to the observation and a modeled production rate adjusted to match the derived column density .
the latter was based on fluorescence efficiencies derived by @xcite .
with the availability of detailed solar lyman-@xmath9 fluxes and line shapes from soho / sumer @xcite , the heliocentric velocity dependent g - factors were recalculated and the column density revised downward by a factor of about three , which is consistent with the water production rate for the date given by @xcite the results are given in table [ tabq ] .
for the comets in which the non - thermal co is detected , the production rates of relative to that for are generally in the range of values found from infrared and millimeter observations .
@xcite reported on four comets observed at millimeter wavelengths including the three comets observed by _ fuse _ at comparable epochs in 2001 .
for comets c/2001 a2 and c/2000 wm1 , our relative production rates are comparable to those of @xcite based on their model that assumes a distributed source with an effective parent lifetime about a factor of 2 longer than the true photochemical lifetime . with an assumed parent , @xcite derive relative production rates @xmath43 times smaller .
millimeter observations of comet c/2001 q4 were made in mid - may 2004 by @xcite .
their results for , again assuming a distributed grain source , are in close agreement with the _ fuse _ value . while infrared bands of near @xmath44 m have been detected in a number of recent comets , the only data available for comparison with the _ fuse _
observations is of c/2001 a2 @xcite .
@xcite , observing 2 and 3 days prior to the _ fuse _ observation , found a low value of 0.24% for the ratio of the production rate of relative to on july 9.5 followed by a decrease of a factor of 4 one day later .
in contrast , we derive a much higher value for this ratio on july 12.5 . the apparent variability of is discussed below in section [ caveat ] . modeling the brightness is not as straightforward .
the molecules initially carry away most of the excess energy of dissociation as translational energy ( @xmath211 km s@xmath33 ) , but they are rapidly thermalized in the inner coma @xcite .
the excited state populations will also vary with incident wavelength and be altered by collisions , so the actual number of molecules in the @xmath45 level produced by solar photodissociation will have a large uncertainty .
nevertheless , as a consistency check we can evaluate the number of such molecules in the field - of - view needed to produce the observed 0.13 rayleigh brightness of the ( 1,1 ) q(3 ) line at 1031.86 observed in comet c/2001
the fluorescence efficiency , or g - factor , at the peak of the solar line is calculated using the solar minimum irradiance recommended by @xcite and the oscillator strengths and branching ratios taken from the compilation of @xcite .
the derived g - factor is @xmath46 s@xmath33 molecule@xmath33 at 1 au , so that a mean column density of @xmath47 @xmath34 is sufficient to produce the observed emission rate . from table
[ tabn ] , the column density of non - thermal co in this comet is @xmath48 @xmath34 , and 78% of those are accompanied by rather than by two h atoms @xcite .
thus 11% of the total needs to be in @xmath49 , @xmath20 , which is plausible based on the laboratory data of @xcite and @xcite . given all of the uncertainties , this is good agreement and supports our proposal that is the source of both the non - thermal co and the vibrationally excited detected by _
fuse_. the vibrational population
may also be examined with the identification of solar lyman-@xmath5 pumping of lines in the lyman ( 1,2 ) band as described in section [ vib ] .
the solar lyman-@xmath5 spectral flux is roughly two orders of magnitude larger than that of @xmath41031.9 resulting in comparably larger fluorescence efficiencies for these lines .
using solar minimum lyman-@xmath5 fluxes from @xcite and molecular data for the lyman bands from @xcite , the g - factors for the ( 1,1 ) p(5 ) and r(6 ) lines can be calculated .
the relative population of the @xmath50 , @xmath23 , needed to account for the observed 1161.8 emission of 0.24 rayleighs is 2% .
this is consistent with the relative population of this level seen in the laboratory experiments @xcite , when we consider that these experiments are done at discrete wavelengths while in the cometary context photodissociation is produced by the wavelength integrated solar flux .
we conclude that photodissociation of is sufficient to produce vibrationally excited that accounts for the presence of lyman-@xmath5 excited fluorescence of in the coma .
while the production rate of is @xmath20.5% relative to , the solar photodissociation rate into is @xmath2100 times higher for , depending on the solar activity level @xcite .
thus both molecules contribute comparable amounts of to the coma .
the branching of water photodissociation into + o(@xmath40d ) has also been studied in the laboratory @xcite , but there are no available measurements of the resultant level populations .
although there has been extensive theoretical work on the h + oh channel of photodissociation @xcite , there does not seem any devoted to the + o(@xmath40d ) channel , which is clearly warranted .
on the basis of the brightness of the lyman-@xmath9 pumped fluorescent lines , @xcite suggested that solar photodissociation of water led to a thermal population of molecules in the @xmath51 , @xmath52 level .
we noted above the utility of the soho / swan data of @xcite to obtain daily values of the water production rate .
for c/2001 a2 near the time of the _ fuse _ observations , the comet exhibited several visual outbursts , particularly one on 2001 july 11.5 , one day before the _ fuse _
observations , which saw a 1.5 magnitude increase in visual brightness @xcite .
these outbursts were reflected in sudden increases in @xmath42 of factors of 23 in one day in the data of @xcite . the _ fuse _ observations of this comet spanned 7.5 hours during which the co emission decreased by a factor of two ( fig .
[ a2_temporal ] ) , likely a manifestation of the decrease in coma abundance following the outburst .
note that for this comet the _ fuse _
@xmath15 aperture projects to @xmath53 km on the sky so that a molecule moving radially outward at 1 km s@xmath33 would exit the field - of - view in less than one hour .
hence the observed variation represents variability of the co source on this time scale . on the other hand ,
the column densities given in table [ tabn ] are taken from the summed spectra ( to improve the signal / noise ratio ) and thus represent average values .
caution is therefore urged in deriving relative production rates from diverse observations , even if obtained on the same date , and in the comparison of results from different observers .
the july 11.5 outburst makes it difficult to compare the derived production rates of @xcite on july 9.5 and 10.5 , noted above , with the present results . from the limited data available ,
it is not possible to determine whether production was enhanced by the outburst or whether the increase inferred from the _ fuse _ spectra is just the result of intrinsic variability responsible for the change seen by @xcite a few days earlier .
the same holds for a comparison of our results with the co production rate derived from the same infrared spectrum on july 10.5 @xcite even though in this case the values of @xmath54 are comparable .
a similar caveat applies to comet c/2001 q4 which _ fuse _ observed over a 27 h period .
the variation of the co emission is shown in fig .
[ q4_temporal ] and can be fit with a sinusoidal curve of period 17.0 h and 22% amplitude . yet
the co production rate , derived from _ hst_/stis observations of the co fourth positive system @xcite , made near a minimum in the _ fuse _ lightcurve , is consistent with the mean value derived from the _ fuse _ spectra .
in contrast , comet c/2001 wm1 showed little variation in co emission over a 60 h period . for this comet ,
hst_/stis observations were also obtained and @xcite derived comparable co production rates from both sets of spectra .
as noted in section [ electron ] , for molecules with long lifetimes against solar photodissociation such as , electron impact dissociation becomes an important destruction channel . using the electron impact rates given in table [ tabg ] , however , would still require an abnormally large ( @xmath55% ) abundance of to account for the derived `` hot '' co column density .
since there are no other suitable parents for co other than and , it is likely that the model used to calculate the values given in table [ tabg ] significantly underestimates the electron impact contribution to dissociation .
@xcite note that the threshold for this process is at 7.4 ev whereas the channel leading to the @xmath37 state of co requires 11.4 ev . at low electron energies , the model of @xcite
does not match the data of @xcite , which appears to follow an @xmath56 distribution , so that there may be an additional large contribution to the dissociation rate from low energy electrons .
there are also uncertainties in the radial and angular distribution of the electrons that make our estimate quantitatively unreliable .
a more complete analysis of cometary photoelectrons , which is beyond the scope of the present paper , would entail an analysis of co cameron band emission similar to that presented by @xcite .
fortunately , relevant _
hst_/stis spectra do exist for comet c/2001 q4 and will be treated in a separate publication .
the first evidence for a distributed source of co in the coma of a comet came from the _ in situ _ neutral mass spectrometer experiment on the _ giotto _ mission to comet 1p / halley @xcite .
@xcite also showed , from the same data set , that the also appeared to come from an extended source and that the was sufficient to account for the extended co @xcite .
polyoxymethylene ( pom ) , evaporated from organic material on refractory grains , has been postulated as the source of this @xcite . in recent years ,
as high resolution spectroscopic instrumentation in the near infrared has made the observation of co in comets almost routine , considerable effort has been expended in trying to determine if such distributed sources are common or unique to certain comets @xcite .
identification of a distributed emission source is based on long - slit spectroscopy in which the spatial profile of the emission deviates from that expected for radial outflow @xcite .
for co this can be done in both the infrared and the ultraviolet .
this approach is applicable only to the brightest comets because the spatial pixels along the slit must be small enough to provide adequate spatial resolution . for such comets , such as c/1995 o1 ( hale - bopp ) ,
the co column densities along the line - of - sight ( to both the sun and to the observer ) are sufficiently large that optical depth effects tend to produce a flattening of the spatial profile near the nucleus that mimics the effect of a distributed source ( the ultraviolet case has been discussed by * ? ? ?
this concern was addressed by @xcite in their analysis of co in comet hale - bopp .
they found a very high production rate of co relative to that of water , 24.1% , with half being from the nucleus and the other half from a distributed source . in the same comet
, @xcite found an even higher relative abundance of co , 3741% , with 90% coming from a distributed source .
@xcite questioned both of these results suggesting that they were the effect of a large infrared opacity , even though both @xcite and @xcite accounted for opacity effects . in any case
, if such a distributed source were it would imply an abundance in this comet an order of magnitude higher than that cited by @xcite .
spectroscopic observations of four comets by _ fuse _ have opened a new window for the investigation of the physics and chemistry of the cometary coma .
the high spectral resolution of _ fuse _ has made it possible to separate the multiple components of the co emission , while the high sensitivity has allowed the detection of fluorescent emission pumped by solar radiation . both of these can be understood in terms of the dissociation processes of and . despite the small sample ,
it is clear that previous ultraviolet spectroscopic observations of co in comets including those from sounding rockets , _ iue _ and _ hst _ ( e.g. , * ? ? ?
* ) probably included these additional components of co that were not resolved , either spectrally or spatially , from the native ( `` cold '' ) co with the instrumentation used .
photoelectron excitation contributes @xmath210% of the observed emission of the co @xmath16 ( 0,0 ) band and needs to be accounted for in deriving co production rates from _ fuse
_ observations .
we thank the _ fuse _ ground system personnel , particularly b. roberts , t. ake , a. berman , b. gawne , and j. andersen , for their efforts in planning and executing these moving target observations .
this work is partially based on data obtained for the guaranteed time team by the nasa - cnes - csa _ fuse _ mission operated by the johns hopkins university .
financial support was provided by nasa contract nas5 - 32985 and nasa grant nag5 - 12963 .
@lcccc@ comet & c/1999 t1 & c/2001 a2 & c/2000 wm1 & c/2001 q4 + & ( mcnaught - hartley ) & ( linear ) & ( linear ) & ( neat ) + date & 2001 feb 3 & 2001 jul 12 & 2001 dec 710 & 2004 apr 24 + start time ( ut ) & 17:43 & 13:48 & 08:48 & 00:40 + exposure duration ( s ) & 15,614 & 15,063 & 36,467 & 68,282 + @xmath57 ( au ) & 1.43 & 1.20 & 1.12 & 1.03 + @xmath58 ( au ) & 1.29 & 0.30 & 0.34 & 0.51 + @xmath59 ( km s@xmath33 ) & 14.9 & 22.8 & 28.3 & 10.8 + @xmath60 ( sfu ) & 160 & 140 & 220 & 115 + @xmath16 ( 0,0 ) ( rayleighs ) & @xmath61 & @xmath62 & @xmath63 & @xmath64 + @xmath65p(40 ) peak ( rayleighs ) & @xmath66 & @xmath67 & @xmath68 & @xmath69 + @xmath29 ( 0,0 ) ( rayleighs ) & @xmath70 & @xmath71 & @xmath72 & @xmath73 + @lcccc@ comet & c/1999 t1 & c/2001 a2 & c/2000 wm1 & c/2001 q4 + & ( mcnaught - hartley ) & ( linear ) & ( linear ) & ( neat ) + @xmath74 ( @xmath34 ) & @xmath75 & @xmath76 & @xmath77 & @xmath78 + @xmath79 ( @xmath34 ) & @xmath80 & @xmath81 & @xmath82 & @xmath83 + @xmath84 ( @xmath34 ) & @xmath85 & @xmath86 & @xmath87 & @xmath88 + @xmath89 ( k ) & @xmath90 & @xmath91 & @xmath92 & @xmath93 + @xmath94 ( k ) & 500 & 500 & 500 & 600 + @lcccc@ comet & c/1999 t1 & c/2001 a2 & c/2000 wm1 & c/2001 q4 + & ( mcnaught - hartley ) & ( linear ) & ( linear ) & ( neat ) + @xmath29 ( 0,0 ) & @xmath95 & @xmath96 & @xmath97 & @xmath98 + @xmath16 ( 0,0 ) & @xmath99 & @xmath100 & @xmath101 & @xmath102 + @xmath103 ( all bands ) & @xmath104 & @xmath105 & @xmath106 & @xmath107 + total dissociation & @xmath108 & @xmath109 & @xmath110 & @xmath111 + @lcccc@ comet & c/1999 t1 & c/2001 a2 & c/2000 wm1 & c/2001 q4 + & ( mcnaught - hartley ) & ( linear ) & ( linear ) & ( neat ) + @xmath54 ( s@xmath33 ) & @xmath112 & @xmath113 & @xmath114 & @xmath115 + @xmath116 ( s@xmath33 ) & @xmath117 & @xmath118 & @xmath119 & @xmath120 + @xmath42 ( s@xmath33 ) & @xmath121 & @xmath122 & @xmath123 & @xmath124 + @xmath54/@xmath42 ( % ) & 10.3 & 2.2 & 0.41 & 7.2 + @xmath116/@xmath42 ( % ) & @xmath1250.5 & 1.3 & 0.52 & 0.40 + | observations of four comets made with the _ far ultraviolet spectroscopic explorer _ show the rotational envelope of the ( 0,0 ) band of the co hopfield - birge system ( @xmath0 @xmath1 ) at 1088 to consist of both `` cold '' and `` hot '' components , the `` cold '' component accounting for @xmath275% of the flux and with a rotational temperature in the range 5575 k. we identify the `` hot '' component as coming from the dissociation of into rotationally `` hot '' co , with electron impact dissociation probably dominant over photodissociation near the nucleus .
an additional weak , broad satellite band is seen centered near the position of the p(40 ) line that we attribute to co fluorescence from a non - thermal high @xmath3 rotational population produced by photodissociation of formaldehyde into co and .
this process also leaves the preferentially populated in excited vibrational levels which are identified by fluorescent lines in the spectrum excited by solar @xmath41031.9 and solar lyman-@xmath5 .
the amount of produced by dissociation is comparable to the amount produced by photodissociation of .
electron impact excitation of co , rather than resonance fluorescence , appears to be the primary source of the observed @xmath6 @xmath1(0,0 ) band at 1151 . |
investigation of reactions involving weakly bound projectiles and the influence of their low binding energies on various reaction channels has received a fillip in recent years , especially in the context of the increasing number of radioactive ion beam facilities . to have a proper understanding of the influence of breakup of loosely bound projectiles on the fusion process , one needs to understand the mechanisms of all the competing reaction channels .
measurements involving weakly bound projectiles , both stable and unstable , with @xmath0+x cluster structures show substantially large production cross sections for @xmath0-particles [ 1 - 9 ] , which indicate the presence of mechanisms other than the @xmath0+x breakup .
_ showed that for the reaction @xmath5li+@xmath2 tb @xcite , about half of the @xmath0 and triton yield originates from the breakup - fusion process , which is more commonly referred to as the incomplete fusion ( icf ) process .
evidence of transfer - induced breakup producing @xmath0-particles in the reaction @xmath5li+@xmath6cu has also been reported @xcite .
our recent works on the systematic measurements of complete and incomplete fusion excitation functions for the reactions @xmath7li+@xmath2 tb and @xmath8b+@xmath2 tb @xcite have shown that the complete fusion ( cf ) cross sections at above - barrier energies are suppressed for reactions with weakly bound projectiles , and the extent of suppression is correlated with the @xmath0-breakup threshold of the projectile .
the measurements also showed that the @xmath0-emitting channel is the favoured icf process in reactions with projectiles having low @xmath0-breakup thresholds .
a critical insight into these measurements shows that the sum of the cf and the icf cross sections for each system yields the total fusion cross sections which lie very close to the calculated one dimensional barrier penetration model calculations , at energies above the barrier .
this shows that the suppression in the cf cross sections at above - barrier energies is primarily due to the loss of flux into the icf channel . a recent exclusive measurement on the reaction @xmath1li+@xmath9pb @xcite showed that the cross sections of the breakup process following @xmath4-stripping ( transfer - breakup ) of @xmath1li are higher than that for the breakup of @xmath1li into @xmath0 and _ d _ fragments .
by contrast , another recent work on the reaction @xmath1li+@xmath10bi @xcite , aimed at disentangling the reaction mechanisms responsible for the large inclusive @xmath0-particle cross sections , indicated that the cross sections of the breakup of @xmath1li into @xmath0 and _ d _ fragments are much higher than those of the breakup following @xmath4- stripping of @xmath1li .
however , very recently it has been reported @xcite that for @xmath7li induced reactions with @xmath11pb and @xmath10bi targets , projectile breakup is triggered predominantly by nucleon transfer , @xmath12-stripping for @xmath1li and @xmath13-pickup for @xmath5li .
based on the observations made in a few reactions , it will perhaps be too optimistic to generalize the dominance of transfer induced breakup for all @xmath7li induced reactions , as the importance of a transfer reaction depends largely on the projectile - target combination . to conclude whether the observation is
a general feature of @xmath7li induced reactions or is true only for specific reactions , it is important to carry out a systematic investigation of @xmath7li induced reactions on various targets , especially medium and light mass targets . in the background of this scenario we chose to carry out an inclusive measurement of the @xmath0-particles produced at energies around the coulomb barrier in the @xmath1li induced reaction with a @xmath2 tb target .
the reaction was so chosen because detailed cf and icf cross sections have already been measured for the system @xcite .
the @xmath1li beam with energies @xmath14=23 , 25 , 27 , 30 and 35 mev , from the 14ud barc - tifr pelletron accelerator centre in mumbai , was used to impinge a self - supporting @xmath2 tb target foil of thickness @xmath15450 @xmath16g/@xmath17 .
the beam energies were corrected for loss of energy in the target material at half - thickness of the target . to detect and identify the @xmath0-particles produced in the reaction , four @xmath18e - e telescopes of si - surface barrier detectors
were placed on a movable arm inside a scattering chamber of 1 m diameter .
the thicknesses of the detectors were so chosen that the @xmath0-particles lose part of their kinetic energies in the first detector ( @xmath18e ) and are stopped in the second detector ( e@xmath19 ) .
the @xmath0-particles produced in the reaction were measured in the range 30@xmath20 @xmath21165@xmath20 in steps of 2@xmath20 or 5@xmath20 depending on the bombarding energy , where @xmath22 is the scattering angle in laboratory .
two si - surface barrier detectors , each of thickness 500 @xmath16 m , were placed at angles of @xmath23 20@xmath20 with - respect - to the beam direction for beam monitoring and normalization purposes .
figure 1(a ) shows a typical two - dimensional inclusive @xmath18e@xmath24e ( e=@xmath18e+e@xmath19 ) spectrum taken at the laboratory scattering angle , @xmath22=99.5@xmath20 for a beam energy of 27 mev . the enclosed area in the figure shows the @xmath0-particle band and its one - dimensional projection
is shown in fig.1(b ) .
it shows a broad continuous peak , with centroid nearly equal to 2/3 times the incident beam energy .
the contribution of the @xmath0-particles , emitted mostly at energies corresponding to the beam velocity , is expected to originate from breakup related processes .
it needs to be mentioned here that the heavy compound nuclei formed , following either the cf or icf process , are expected to decay predominantly by neutron evaporation @xcite and this is also predicted by the statistical model calculations done using the code pace2 @xcite .
( color online ) a ) typical two - dimensional @xmath18e - e spectrum of the reaction @xmath1li+@xmath2 tb at scattering angle , @xmath25=99.5@xmath20 for the beam energy 27 mev .
the enclosed area shows the @xmath0-particle band .
b ) the one - dimensional projection of the enclosed area in the upper figure ( a ) . ] the differential cross sections of the inclusive @xmath0-particles were obtained by using the formula , @xmath26 where @xmath27 and @xmath28 are the number of counts under the broad continuous peak of the @xmath0-particles ( fig.1 ) and the average number of counts in the monitor detectors , respectively .
the quantities @xmath29 and @xmath30 are the solid angles subtended by the monitor detectors and the @xmath18e - e telescope , respectively and @xmath31 is the angle of the monitor detector .
for all the five bombarding energies , the broad peak in each of the @xmath0-particle energy spectra was well separated from the low - energy small peak ( fig.1 ) , at all scattering angles .
the @xmath0-particles in the low - energy peak , which is indeed a very small contribution at all the bombarding energies , could be due to target impurities , such as c and o. the measured angular distributions of the inclusive @xmath0-particles for the five incident energies are shown in fig.2 .
the angular distribution at each of the bombarding energies was obtained by considering the counts within the main peak of the @xmath0-spectrum . with the exception of the low energy 23 mev data , each of the distributions shows a clear maximum that shifts to lower laboratory angle with the increase of beam energy .
the angular distribution data were fitted with gaussian functions and are shown by the lines in fig.2 .
the total angle - integrated @xmath0-particle cross sections obtained from the angular distribution data at each of the incident energies are plotted in fig.3 .
( color online ) angular distributions of inclusive @xmath0-particles for the reaction @xmath1li+@xmath2 tb at energies @xmath14=23 - 35 mev .
the lines through the data are fits with gaussian functions .
] ( color online ) measured inclusive @xmath0-particle cross sections for the reaction @xmath1li+@xmath2 tb .
the dash - dotted line shows the ncbu cross sections obtained from the cdcc calculations . ]
because the present work is an inclusive measurement , the @xmath0-particle cross sections are expected to have contributions from various processes . for reactions induced by the weakly bound projectile @xmath1li ( q = + 1.47 mev for the @xmath0+@xmath3 breakup )
, it is natural to assume that an important contributor to the @xmath0-particle cross sections is the breakup of @xmath1li into @xmath0 and @xmath3 fragments .
besides , other processes producing significant @xmath0-particle cross sections are also likely to occur . the processes that might contribute significantly to the inclusive @xmath0-particle cross sections are : + i ) breakup of @xmath1li into @xmath0and @xmath3 fragments , which could be either direct or resonant ( i.e. sequential ) or both , where both fragments escape without being captured by the target , i.e. , a no - capture breakup ( ncbu ) process , + ii ) @xmath0-particles resulting from @xmath3-capture by the target ( @xmath3-icf ) , following the breakup of @xmath1li into @xmath0 and @xmath3 , or a one - step @xmath3- transfer to the target , + iii ) single - proton stripping from @xmath1li to produce unbound @xmath32he that decays to an @xmath0-particle plus a neutron , + iv ) single - neutron stripping from @xmath1li to produce @xmath0-unstable @xmath32li , that will subsequently decay to an @xmath0-particle plus a proton , and + v ) single - neutron pickup by @xmath1li to produce @xmath5li , which breaks into an @xmath0-particleand a triton if @xmath5li is excited above its breakup threshold of 2.45 mev .
in order to understand the origin of the large inclusive @xmath0-particle cross sections obtained in the reaction @xmath1li+@xmath2 tb , measurements and/or theoretical calculations are necessary to estimate the contribution from each of the above processes . to estimate the contribution from the ncbu process ( i ) , exclusive measurements between the breakup fragments @xmath0 and @xmath3 are needed .
as only inclusive measurements were taken in the present work , the ncbu cross sections have been estimated theoretically in the framework of the continuum - discretized - coupled channels ( cdcc ) method @xcite .
the cdcc calculations were performed with the coupled channels code fresco @xcite ( version frxx.09j ) , by assuming @xmath1li to have an @xmath0+@xmath3 cluster structure for its bound and continuum states .
following ref.@xcite , the @xmath0-@xmath3 continuum was discretized into a series of equally spaced momentum bins , each of width @xmath18k = 0.25 @xmath33 in the range 0.0 @xmath34 k @xmath340.75 fm @xmath35 , corresponding to the @xmath1li excitation energy of 1.47 @xmath34e@xmath36 @xmath3410.27 mev with respect to the @xmath1li ground state energy .
the contribution from higher excited states is expected to be negligible .
each momentum bin was treated as an excited state of @xmath1li nucleus with excitation energy equal to the mean energy of the bin and having spin j and parity ( -1)@xmath37 . the angular momenta are related by * j*=*l*+*s * , where * s * is the spin of the @xmath3 and * l * is the relative angular momentum of the @xmath0@xmath24@xmath3 cluster system . in the calculations ,
l is limited to 0 , 1 , and 2 .
the contribution from higher l is negligible .
couplings to the 3@xmath38 ( e@xmath39= 2.18 mev ) , 2@xmath38 ( e@xmath39=4.31 mev ) and 1@xmath38 ( e@xmath39=5.65 mev ) resonant states as well as couplings to the non - resonant @xmath0+@xmath3 continuum were included in the calculations . in order to avoid double counting ,
the bin width was suitably modified in the presence of resonant states .
the @xmath0+@xmath3 binding potentials were taken from ref.@xcite .
the cluster - folding model potentials for the interactions , @xmath0@xmath24target and @xmath3@xmath24target were evaluated at 2/3 and 1/3 of the incident energy of the @xmath1li beam , respectively . as no experimental elastic scattering angular distribution data for @xmath0+@xmath2 tb and @xmath3+@xmath2 tb reactions are available in the literature , the global optical model potential parameters @xcite were used in describing the interactions at the corresponding energies .
the couplings from the ground state to continuum and continuum to continuum states were included in the calculations .
both coulomb and nuclear couplings were incorporated .
the results of the ncbu cross sections thereby calculated are plotted in fig.3 by the dash - dotted curve , and they are seen to largely underestimate the measured @xmath0-particle cross sections .
this shows that the @xmath0-particles from sources other than breakup are important and need to be accounted for .
this feature has also been observed for other heavy systems , such as @xmath7li+@xmath9pb @xcite and @xmath1li+@xmath10bi @xcite .
the @xmath0-particle cross sections resulting from the @xmath3-capture by the @xmath2 tb target ( @xmath3-icf , process ( ii ) ) , followed by @xmath40 evaporation , were determined from the @xmath41-ray spectra recorded in the fusion cross sections measurement of the @xmath1li+@xmath2 tb reaction @xcite .
the cross sections of the resulting residual nuclei @xmath42dy , @xmath2dy , and @xmath43dy were already reported in ref .
however , for the sake of convenience , the cross sections of the @xmath44 channels , following the @xmath3-capture icf , along with the total @xmath3-capture cross sections ( @xmath45@xmath44 ) are plotted in fig.4 . as already mentioned in the earlier work
, the icf cross sections thus measured also include contributions due to the @xmath3-transfer from @xmath1li to @xmath2 tb , if any , since in the @xmath41-ray measurement it was not possible to distinguish between the two types of events .
also , the single - proton stripping process ( iii ) @xmath2tb(@xmath1li,@xmath32he)@xmath42dy ( q = + 2.836 mev ) , if it occurs , will lead to the @xmath42dy nuclei in excited states .
the @xmath42dy nuclei following the @xmath46-stripping process will then decay by @xmath40 evaporation to produce dy - isotopes and will be included in the @xmath47 channel cross sections of the @xmath3-icf process .
( color online ) cross sections of @xmath42dy , @xmath2dy and @xmath43dy nuclei produced in the reaction @xmath1li+@xmath2 tb .
the hollow points are the sum of the cross sections of the three dy isotopes , i.e. the total @xmath3-icf cross sections for the reaction .
see sec.3.2 for details . ] the contributions due to the processes ( iv ) and ( v ) , i.e. the single - neutron stripping reaction @xmath2 tb ( @xmath1li,@xmath32li)@xmath42 tb ( q = + 0.711 mev ) , and the single - neutron pickup reaction @xmath2tb(@xmath1li,@xmath5li)@xmath43 tb ( q = + 0.883 mev ) , leading to the @xmath41-rays of @xmath42 tb and @xmath43 tb respectively , have been measured from the @xmath41-ray spectra @xcite of the @xmath1li+@xmath2 tb reaction . for the former case , the production cross section of the 63.68 kev ( 1@xmath48 ) state of @xmath42 tb was obtained from the measured cross section of the 63.68 kev @xmath41-ray , after correcting for its internal conversion coefficient ( @xmath49 ) of 15.1 . in the @xmath41-ray spectra , this was the only @xmath41-ray of @xmath42 tb that could be identified . besides , because this is a fairly low energy @xmath41-ray , special care was taken to estimate the area under this @xmath41-ray peak .
the 63.68 kev @xmath41-ray is an e2 transition that feeds the ground state of @xmath42 tb . as this @xmath41-ray has a fairly large internal conversion coefficient , the reliability of the cross sections of @xmath42 tb
may be questioned . the large value of @xmath49 = 15.1 ,
though theoretically calculated , is expected to be a reliable estimate since theoretically calculated values of @xmath49 , especially for e2 transitions , are known to agree well with the experimentally measured values . for example
, the measured value of @xmath49 for the 75.26 kev transition in @xmath42gd is [email protected] while the calculated value varies between 7.24 and 7.51 ; the measured @xmath49 value for the 73.39 kev transition for @xmath50dy is [email protected] while the calculated value varies between 8.80 and 9.12 ; the measured @xmath49 value for the 53.2 kev transition for @xmath51th is 229@xmath237 and the calculated value lies between 227.6 and 234.2 @xcite . also , although the cross section of the 63.68 kev @xmath41-ray is small , the corresponding peak in the @xmath41-spectrum is fairly clean , thereby yielding @xmath41-ray cross section with small uncertainty .
nevertheless , an uncertainty of 10@xmath52 in the theoretical value of @xmath49 has been assumed while obtaining the cross sections of the @xmath42 tb nuclei shown in fig.5 .
it should be emphasized here that in this method of extraction of @xmath12-stripping cross sections from the @xmath41-ray spectra , the contribution of transfer to the ground state of @xmath42 tb can not be determined .
so within the constraints of the present technique , only excited state transfer cross sections could be obtained .
( color online ) measured cross sections of @xmath4-stripping and @xmath4-pickup processes populating the excited states of @xmath42 tb and @xmath43 tb nuclei respectively , in the reaction @xmath1li+@xmath2 tb .
the dashed and solid curves are the dwba calculated cross sections , with spectroscopic factors of 1 and 0.25 , respectively , assumed for the 63.68 kev state . ] similarly , the total cross sections of the @xmath4-pickup process populating the excited states of @xmath43 tb nuclei were obtained by summing the measured cross sections of the 162.2 kev and 89.08 kev @xmath41-rays after appropriate correction for their respective internal conversion factors , and these are shown in fig.5 . in this case also , the ground state transfer cross sections could not be determined . in order to compare the measured cross sections for the @xmath12-stripping process with theory
, we attempted to calculate the cross sections for the single @xmath12-transfer to the first excited 63.68 kev ( 1@xmath53 ) state of @xmath42 tb nuclei for the five bombarding energies of 23 , 25 , 27 , 30 and 35 mev .
the transfer cross sections were calculated in the distorted wave born approximation ( dwba ) framework , using the computer code fresco .
the @xmath12@xmath24@xmath2 tb and @xmath12@xmath24@xmath32li binding potentials were taken from refs . @xcite and @xcite respectively .
the required potential parameters for the entrance channel @xmath1li+@xmath2 tb , the exit channel @xmath32li+@xmath42 tb and the @xmath32li+@xmath2 tb core - core interaction were taken to be the global optical model potential parameters of ref .
@xcite with modifications such that these potentials fit the measured elastic scattering angular distributions at the five bombarding energies of 23 , 25 , 27 , 30 and 35 mev .
depth parameters have been adjusted to reproduce the binding energy of the neutron to the core @xmath2 tb .
the spectroscopic factor ( sf ) for @xmath1li@xmath54@xmath32li + @xmath12 was taken from ref .
the experimental sf for the 63.68 kev ( 1@xmath53 ) state in @xmath42 tb is not available in the literature .
nevertheless , the transfer cross sections have been calculated by assuming the sf to be 1.0 for the 63.68 kev ( 1@xmath53 ) state .
the cross sections thereby calculated are shown by the dashed curve in fig.5 , and they are seen to largely over - predict the measured cross sections .
it was found that the dwba calculations done with a sf of 0.25 gave an overall fit to the measured cross sections at the higher energies .
the resulting calculations are shown in the figure by the solid curve . due to the unavailability of relevant sfs , no better dwba calculation could be done for the @xmath4-stripping process .
therefore , no further attempt was undertaken to calculate the cross sections for the @xmath4-pickup process .
( color online ) contributions to the @xmath0-particle cross sections originating from various processes in the reaction @xmath1li+@xmath2 tb .
the squares are the measured inclusive @xmath0-particle cross sections .
the up triangles are the @xmath45@xmath44 channels , corresponding to the @xmath3-icf process ( including @xmath3-transfer and @xmath46-stripping , if any ) .
the down triangles are the contributions from the @xmath4-stripping process ( excluding ground state transfer ) , corresponding to the instantaneous decay of the resulting @xmath0-unstable @xmath32li nuclei into @xmath55 and @xmath13 .
the circles are the sum of the @xmath3-icf and the @xmath4-stripping cross sections .
the cross sections resulting from the breakup of @xmath1li into @xmath0 and @xmath3 ( ncbu process ) , as determined from the cdcc calculations are shown by the dash - dotted line .
the dotted curve shows the trend of the cross sections , if the calculated ncbu cross sections are added to the measured cross sections of @xmath3-capture icf and @xmath4-stripping processes . ] the measured @xmath3-icf ( i.e.@xmath56 ) cross sections , the @xmath4-stripping cross sections ( excluding the ground state transfer contribution ) and the sum of the cross sections from these two processes are compared with the measured inclusive @xmath0-particle cross sections in fig.6 .
the calculated ncbu cross sections are also shown in the figure by the dash - dotted line .
it is observed that the @xmath4-stripping cross sections ( excluding the ground state transfer contribution ) are much larger than the calculated ncbu cross sections , in contradiction to that reported for the @xmath1li+@xmath10bi reaction @xcite but in agreement with the observation of luong _ et al .
it had been mentioned in ref .
@xcite that the measured exclusive cross sections of @xmath0+@xmath13 in the @xmath1li+@xmath10bi reaction , following @xmath12-stripping of @xmath1li , are possibly the lower limit .
this may be because the detector configuration used to measure the @xmath0+@xmath13 breakup cross section @xcite did not cover the whole range of the relative momentum , thereby leading to the underestimation of the cross section . though the relative importance of reaction mechanisms largely depends on the target - projectile combination , the present observations , in conjunction with those reported in refs .
@xcite , in fact do show that the @xmath12-stripping process is more important than the ncbu process in @xmath1li - induced reactions with targets such as @xmath2 tb , @xmath11pb and @xmath10bi . it can be seen from fig.6 that the sum of the cross sections resulting from @xmath3-icf ( including @xmath3-transfer and @xmath46-stripping , in any ) and @xmath4-stripping ( excluding ground state transfer ) reactions , shown by the solid circles , lie very close to the measured total inclusive @xmath0-particle cross sections . here
the yield of the @xmath0-particles due to the ground state transfer in the @xmath4-stripping process , and also following the breakup of @xmath5li nuclei produced via the @xmath12-pickup process have not been considered .
the @xmath0-breakup threshold of @xmath5li is 2.45 mev , and hence @xmath5li nuclei can breakup only if they are excited above 2.45 mev .
therefore , the @xmath4-pickup reaction will contribute to the total @xmath0-particle cross sections , depending on the excitation energy of the @xmath5li nuclei . at lower bombarding energies
, this process may not be a significant contributor .
but at higher bombarding energies , the @xmath5li nuclei may be excited to energies above the breakup threshold , thereby resulting in a small contribution
. however , it is obvious from the figure that for this reaction , over the energy range of the present measurement , the @xmath4-pickup process is certainly not a very significant contributor to the total @xmath0-particle cross sections .
the dotted curve in the figure shows that if we add the cdcc calculated ncbu cross sections to the measured @xmath3-icf and @xmath4-stripping cross sections , the inclusive @xmath0-particle cross sections are nearly reproduced .
thus , the @xmath3-icf ( including @xmath3-transfer and @xmath46-stripping , if any ) and the @xmath4-stripping processes are the dominant contributors , with the ncbu process being a relatively small contributor , to the total @xmath0-particle cross sections in the @xmath1li+@xmath2 tb reaction at energies around the coulomb barrier .
in summary , the inclusive @xmath0-particle cross sections for the reaction @xmath1li+@xmath2 tb have been measured at energies around the coulomb barrier . the ncbu cross sections calculated using the cdcc formalism
are found to be only a small fraction of the inclusive @xmath0-particle cross sections .
other reaction mechanisms contributing to the large @xmath0-particle cross sections have been disentangled , using data from our earlier work @xcite based on an entirely different technique , _
e.g. _ the @xmath41-ray method .
the @xmath4- stripping cross sections are found to be much larger than the calculated cross sections of the ncbu process , in contradiction to the observation reported for the reaction @xmath1li+@xmath10bi @xcite .
the @xmath3-icf , including @xmath3-transfer and @xmath13-stripping if any , and the @xmath4-stripping processes are found to be the dominant contributors to the total @xmath0-particle cross sections in the @xmath1li+@xmath2 tb reaction . however , due to the lack of appropriate spectroscopic factors , proper dwba calculations could not be performed .
experiments aimed at measuring such spectroscopic factors need to be carried out in the near future .
besides , as transfer induced breakup seems to be an important process in reactions with loosely bound projectiles , both inclusive and exclusive measurements in other systems , especially lighter systems and systems involving halo and skin nuclei , would be very valuable to obtain a clear picture of the transfer - breakup process .
identification and subsequent determination of the absolute cross sections of different multi - step reaction processes involved in reactions with weakly bound nuclei may pave the way for the theorists to come up with a proper theoretical description of such processes , which is indeed a challenging task .
we are grateful to prof .
b.k . dasmahapatra for valuable discussions and advices at various stages of the work .
we thank mr .
das for his earnest technical help during the experiment .
we would also like to thank the accelerator staff at the barc - tifr pelletron facility , mumbai , for their untiring efforts in delivering the beams .
aguilera , j.j .
kolata , f.m .
nunes , f.d .
becchetti , p.a.de young , m. goupell , v. guimaraes , b. hughey , m.y .
lee , d. lizcano , _ et al .
lett . * 84 * , 5058 ( 2000 ) r. ost , e. speth , k.o .
pfeiffer , and k. bethge , phys .
c * 5 * , 1835 ( 1972 ) k.o .
pfeiffer , e. speth , and k. bethge , nucl .
phys . a * 206 * , 545 ( 1973 ) c.m .
castaneda , h.a .
smith , jr .
singh , j. jastrzebski , h. karwowski , and a.k .
gaigalas , phys .
b * 77 * , 371 ( 1978 ) c.m .
castaneda , h.a .
smith , jr .
singh , and h. karwowski , phys .
c * 21 * , 179 ( 1980 ) p.n .
de faria , r. lichtenthaler , k.c.c .
pires , a.m. moro , a. lepine - szily , v. gumairaes , d.r .
mendes , jr . , a.arazi , a. barioni , v. morcelle , _ et al .
c * 82 * , 034602 ( 2010 ) g.r .
kelly , n.j .
davis , r.p .
ward , b.r .
fulton , g. tungate , n. keeley , k. rusek , e.e .
bartosz , p.d .
cathers , d.d .
caussyn , _ et al .
c * 63 * , 024601 ( 2000 ) c. signorini , a. edifizi , m. mazzocco , m. lunardon , d. fabris , a. vitturi , p. scopel , f. soramel , l. stroe , g. prete , _ et al .
_ , phys . rev .
c * 67 * , 044607 ( 2003 ) h. utsunomiya , s. kubono , m.h .
tanaka , m. sugitani , k. morita , t. nomura , and y. hamajima , phys .
c * 28 * , 1975 ( 1983 ) a. shrivastava , a. navin , n. keeley , k. mahata , k. ramachandran , v. nanal , v.v .
parkar , a. chatterjee , and s. kailas , phys .
b * 633 * , 463 ( 2006 ) a. mukherjee , s. roy , m.k .
pradhan , m.s .
sarkar , p. basu , b. dasmahapatra , t. bhattacharya , s. bhattacharya , s.k .
basu , a. chatterjee , _ et al .
b * 636 * , 91 ( 2006 ) a. mukherjee and m.k .
pradhan , pramana * 75 * , 99 ( 2010 ) m.k .
pradhan , a. mukherjee , p. basu , a. goswami , r. kshetri , s. roy , p.r .
chowdhury , m.s .
sarkar , r. palit , v.v .
parkar , _ et al .
c * 83 * , 064606 ( 2011 ) d.h .
luong , m. dasgupta , d.j .
hinde , r. du rietz , r. rafiei , c.j .
lin , m. evers , and a. diaz - torres , phys .
lett b * 695 * , 105 ( 2011 ) s. santra , s. kailas , v.v .
parkar , k. ramachandran , v. jha , a. chatterjee , p.k .
rath , and a. parihari , , phys . rev .
c * 85 * , 014612 ( 2012 ) d.h .
luong , m. dasgupta , d.j .
hinde , r. du rietz , r. rafiei , c.j .
lin , m. evers , and a. diaz - torres , phys .
c * 88 * , 034609 ( 2013 ) a. gavron , phys .
c * 21 * , 230 ( 1980 ) m. yahiro , n. nakano , y. iseri , m. kamimura , prog .
. suppl . * 89 * , 32 ( 1986 ) n. austern , y. iseri , m. kamimura , m. kawai , g. rawitscher , m. yahiro , phys . rep . *
154 * , 125 ( 1987 ) i.j .
thompson , comput .
phys . rep .
* 7 * , 167 ( 1988 ) n. keeley and k. rusek , phys .
b * 427 * , 1 ( 1998 ) k.i .
kubo and m. hirata , nucl.phys . a * 187 * , 186 ( 1972 ) v. avrigeanu , p.e .
hodgson and m. avrigeanu , phys .
c * 49 * , 2136 ( 1994 ) h. an and c. cai , phys .
c * 73 * , 054605 ( 2006 ) s. raman , c.w .
nestor , jr .
, a. ichihara , and m.b .
trzhaskovskaya , phys .
c * 66 * , 044312 ( 2002 ) f.d .
becchetti , jr . and g.w .
greenlees , phys . rev . * 182 * , 1190 ( 1969 ) s. cohen and d. kurath , nucl
. phys . a * 101 * , 1 ( 1967 ) j. cook , nucl .
a * 388 * , 153 ( 1982 ) s. santra , v.v .
parkar , k. ramachandran , u.k .
pal , a. shrivastava , b.j .
roy , b.k .
nayak , a. chatterjee , r.k .
chowdhury and s. kailas , phys .
b * 677 * , 139 ( 2009 ) | the inclusive cross sections of the @xmath0-particles produced in the reaction @xmath1li+@xmath2 tb have been measured at energies around the coulomb barrier .
the measured cross sections are found to be orders of magnitude larger than the calculated cross sections of @xmath1li breaking into @xmath0 and @xmath3 fragments , thus indicating contributions from other processes .
the experimental cross sections of @xmath4-stripping and @xmath4-pickup processes have been determined from an entirely different measurement , reported earlier .
apart from incomplete fusion and/ @xmath3-transfer processes , the @xmath4-stripping process is found to be a significant contributor to the inclusive @xmath0-particle cross sections in this reaction . |
it has been believed that hadron matter changes to quark matter at high - density region by way of the `` deconfinement phase transition '' .
unfortunately the deconfinement phase transition have not been well understood up to now , and many authors have studied it by model calculations or by first - principle calculations like lattice qcd .
these studies are now developing , and many exciting results have been reported .
properties of quark matter have been actively studied theoretically in quark - gluon plasma , color superconductivity @xcite or magnetism @xcite , and experimentally in relativistic heavy - ion collision ( rhic ) , hera or early universe and compact stars @xcite .
when we calculate uniform hadron matter ( nuclear matter ) and quark matter at zero temperature separately , by using the mit bag model , we can expect the first - order phase transition as seen in figs .
[ eosbulk ] and [ omebulk ] .
we can see that quark matter is an energetically favorable state at high - density region , @xmath3 ( fig .
[ eosbulk ] ) . as we can see in fig .
[ omebulk ] , the thermodynamic potential of quark matter becomes lower at higher baryon - number chemical potential .
these results suggest the deconfinement phase transition at high densities .
the features of the deconfinement phase transition have not been fully elucidated yet .
we assume here that it is the first order phase transition , and use the bag model for simplicity
. then the thermodynamically forbidden region appears in the equation of state ( eos ) and we can expect the mixed phase , the hadron - quark mixed phase , in some density region , which may exist in inner core region of neutron stars and during the hadronization of high - temperature quark - gluon plasma at rhic experiment .
we have to apply the gibbs conditions ( gc ) to get eos in thermodynamic equilibrium : gc demand chemical equilibrium , pressure valance and thermal equilibrium between two phases ; @xmath4 where @xmath5 and @xmath6 are baryon - number and charge chemical potentials , respectively .
thermal equilibrium is implicitly achieved at @xmath7 .
note that there are two independent chemical potentials in this phase transition .
in such a case the system should be much different from the liquid - vapor phase transition which is described by only one chemical potential . of uniform hadron matter and quark matter as a function of baryon chemical potential @xmath8 .
uniform quark matter is energetically favorable in high @xmath8 region while hadron matter in low @xmath8 region.,width=302 ] of uniform hadron matter and quark matter as a function of baryon chemical potential @xmath8 .
uniform quark matter is energetically favorable in high @xmath8 region while hadron matter in low @xmath8 region.,width=302 ] these gc must be fulfilled in the hadron - quark mixed phase . on the other hand
, the maxwell construction ( mc ) may be very familiar and has been used by many authors to get eos for the first order phase transitions @xcite .
it is well known that mc is a correct prescription to derive eos for the liquid - vapor phase transition . however , glendenning @xcite pointed out that mc is not appropriate for the hadron - quark mixed phase : one of gc about the charge chemical equilibrium , @xmath9 , is not satisfied in mc , since the local charge neutrality is implicitly assumed without imposing this condition .
he emphasized that the local charge neutrality is too restrictive and each hadron or quark phase may have a net charge because only the total charge must be kept neutral .
when we use gc in the bulk calculation , which we explain in detail later , we can see the mixed phase appears in a large density region .
the pressure is not constant as density changes in the mixed phase , while only a constant pressure is obtained from mc as shown in refs .
@xcite . however , the bulk calculation is too simple and heiselberg et al .
@xcite claimed the importance of including the finite - size effects , i.e. , the surface tension at the hadron - quark boundary and the coulomb interaction energy .
they studied the quark droplet immersed in hadron matter and found that it is energetically unfavorable if the surface tension is large enough . on the contrary , however , if the surface tension is not large the mixed phase can exist in some density region .
glendenning and pei @xcite suggested the crystalline structure by a bulk calculation using the small surface tension : one phase is immersed in another phase with various geometrical structures ; `` droplet '' , `` rod '' , `` slab '' , `` tube '' , and `` bubble '' . these are called the structured mixed phases ( smp ) . applying the results based on the bulk calculation to neutron stars
, they suggested that there could develop smp in the core region for several kilo meters in thickness .
r120 mm at a first glance , this view seems to be reasonable and there may appear the mixed phase in a large density region . however , there are still many points to be elucidated about the finite - size effects .
voskresensky et al .
emphasized that the proper treatment of the coulomb interaction is important in the mixed phase @xcite .
first note that there is the relation between chemical potential and the coulomb potential by way of the gauge transformation : chemical potential is not well defined before the gauge fixing .
secondly , the charge chemical equilibrium could be rather satisfied even in mc , once the coulomb potential @xmath10 is incorporated : @xmath11 whereas @xmath12 note that the electron chemical potential @xmath13 and the electron number density can be expressed in terms of the combination , @xmath14 ( see eq . ( 48 ) ) , when the coulomb potential @xmath15 is introduced .
therefore , eqs .
( [ mc1 ] ) , ( [ mc2 ] ) mean that the charge chemical potential is equal , while the electron number is different between the hadron and quark phases .
equation ( [ mc2 ] ) is reduced to @xmath16 in the absence of the coulomb potential , which is the previous claim that the charge chemical equilibrium is apparently violated in mc .
thus we can see that the previous claim means nothing but the difference of the electron number in two phases @xcite .
thirdly , it is important to take into account the coulomb screening effect .
we have to solve the poisson equation consistently with other equations of motion for charged particles .
actually , voskresensky et al . showed smp is mechanically unstable by the coulomb screening effect when the surface tension is not small .
if smp is mechanically unstable , the phase transition should be similar to that in mc .
it can be easily seen that there should occur rearrangement of charged particle densities by the coulomb interaction . on the contrary
such rearrangement modifies the coulomb potential . as a result
the poisson equation becomes highly non - linear and difficult to solve analytically . in their study , a linear approximation ( rpa ) was employed to solve the poisson equation analytically .
it is valid to use the approximation for pointing out the important property of smp , but it may be conceivable that various charge properties like `` global charge neutrality '' for gc , `` local charge neutrality '' for mc , and `` the coulomb screening effect '' have important roles in the mixed phase . therefore ,
solving the poisson equation without any approximation is of much significance .
we have reported a preliminary result for the case of droplet @xcite
. it would be also very interesting to derive eos in our consistent calculation for the hadron - quark matter phase transition .
now we consider two infinite matters separated by a sharp boundary : uniform quark matter and hadron matter .
we consider the quark phase consists of u , d , s quarks and electron , and the hadron phase proton , neutron and electron .
we discard the coulomb interaction in the calculation
. then we can evaluate the total thermodynamic potential @xmath17 : @xmath18 the explicit expressions of @xmath19 for non - uniform matter are given in the next section .
we use them by replacing the space dependent quantities by constants for each uniform matter . introducing the volume fraction of the quark phase @xmath20
, we impose the global charge neutrality : the total charge density vanishes , @xmath21 where @xmath22 and @xmath23 are the net charge densities of the quark and hadron phases , respectively , @xmath24 note that we have now six chemical potentials ; @xmath25 , @xmath26 , @xmath27 , @xmath28 , @xmath29 , @xmath30 .
we first consider @xmath31 equilibrium in each phase and chemical equilibrium at the hadron - quark boundary .
thus all chemical potentials are constant .
the chemical equilibrium conditions then are : @xmath32 in the quark phase , @xmath33 in the hadron phase , and @xmath34 at the hadron - quark interface .
the last condition ( [ chemeqp ] ) can be derived from other four conditions , so that there are left four independent conditions for chemical equilibrium .
therefore , if we give two chemical potentials @xmath8 and @xmath30 by hand , we can determine these four chemical potentials ; @xmath25 , @xmath26 , @xmath27 and @xmath28 from these four equations .
next , we can determine @xmath30 by the global charge neutrality condition ( [ totchcond ] ) .
@xmath20 is still unknown at this point and finally , we find the optimal value of @xmath20 by using one of gc ; @xmath35 , where pressure @xmath36 is given by the thermodynamic relation : @xmath37 .
thus once @xmath8 is given , all other values @xmath38 ( @xmath39 ) and @xmath20 can be obtained .
we show the results of the bulk calculation .
we present the charge densities in fig .
[ chdens ] .
we can see the total charge density is zero , while the quark phase is negatively charged and the hadron phase positively charged .
note that mc always gives a null charge density in each phase due to the local charge neutrality @xcite .
r80 mm figure [ presbulk ] shows pressure versus baryon - number density .
the most important difference between `` maxwell '' and `` bulk gibbs '' is that the pressure given by mc is constant while that given by the bulk calculation with gc is density dependent .
remember that the outstanding feature in these two results comes from that each hadron or quark phase has a net charge in `` bulk gibbs '' , while each phase is neutral in `` maxwell '' .
the phase diagram in the @xmath40 plane is presented in fig .
[ mubmuebulk ] .
we can easily see that the charge chemical potential @xmath41 is much different between uniform hadron matter and quark matter for mc , which means chemical equilibrium condition for @xmath41 is not satisfied in mc .
thus it could be said that mc is not a proper way in the bulk calculation .
glendenning @xcite pointed out this defect in mc and showed another result ( bulk gibbs ) within the bulk calculation .
one can easily see that the mixed phase appears in a wide @xmath8 region in fig .
[ mubmuebulk ] . ,
width=302 ] , width=302 ] after the claim by glendenning , heiselberg et al .
@xcite demonstrated that the finite - size effects may disfavor the mixed phase by extending the bulk calculation to include the coulomb interaction and the surface energy . on the other hand , glendenning and
pei @xcite suggested `` crystalline structures of the mixed phase '' because smp are energetically more favorable than uniform matter even including the surface and coulomb effects .
voskresensky et al
. pointed out the importance of the coulomb screening effect and suggested the usefulness of mc for the mixed phase .
we have seen mc is apparently incorrect because if we apply mc to this mixed phase , one of gc , charge chemical equilibrium , is violated . however , once the coulomb interaction is taken into account , there is still a room to satisfy the condition without spoiling the picture given by mc , as noted in the previous section . since
the coulomb screening effect had been taken into account with an approximation in their study , we may still address a question , `` is the picture derived from mc is completely meaningless when we include the coulomb interaction in a proper way ? '' .
all we have to do next is to solve the poisson equation without any approximation , strictly satisfying gc .
we use the idea of the density functional theory ( dft ) @xcite to study the quark - hadron mixed phase .
we employ here the thomas - fermi approximation , which gives the simple expressions for the energies .
we summarize below some energy expressions with the local density approximation ( lda ) ; the expressions are first derived for uniform system , and then we regard densities as the space - dependent functions .
first , the kinetic energy is simply expressed as @xmath42 , \nonumber\end{aligned}\ ] ] where @xmath43 is the particle mass and the fermi momentum @xmath44 is expressed by the density profile @xmath45 as @xmath46 .
next , we consider the interaction energy of quarks .
we take the first - order contribution which comes from the one - gluon exchange interaction in uniform quark matter , @xmath47{oge.eps2}\end{tabular } = \overline{\psi}(p_2 ) \left ( -ig\gamma_\mu \frac{\lambda_a}{2 } \right ) \psi(p_1 ) d^{\mu\nu } ( p^\prime ) \overline{\psi}(p_4 ) \left ( -ig\gamma_\nu \frac{\lambda_a}{2 } \right ) \psi(p_3).\ ] ] here @xmath48 is the su(3 ) gell - mann matrix and @xmath49 the gluon propagator @xcite . by way of the wick contraction , the hartree and fock terms
are derived : @xmath50{hartree.eps2}\end{tabular}\ ] ] for the hartree term , and @xmath51{fock.eps2}\end{tabular}\ ] ] for the fock term . due to the traceless property of @xmath48 , @xmath52 ,
the hartree term gives a null contribution in the color - singlet quark matter , @xmath53 = 0.\ ] ] therefore only the fock term contributes to the interaction energy , which becomes @xmath54{fock2.eps2}\end{tabular } = -\frac{1}{8}g^2\mathrm{tr}\left(\lambda_a \lambda_a\right ) \int\!\!\!\int \frac{d^4k}{\left(2\pi\right)^4 } \frac{d^4q}{\left(2\pi\right)^4 } \mathrm{tr } \left ( \gamma_\mu g_d ( q ) \gamma_\nu g_d ( k ) \right ) d^{\mu\nu}(q - k ) \\
= -64 \alpha_s \pi \int\!\!\!\int \frac{d^3q}{(2\pi)^3 } \frac{d^3k}{(2\pi)^3 } n_q n_k \frac{1}{4k^0q^0 } \left ( 2m_f^2 -kq \right ) \frac{-1}{\left(k - q\right)^2 } \equiv \epsilon_{\mathrm{fock}}.\end{aligned}\ ] ] here @xmath55 is the density - dependent quark propagator @xcite .
the subscript @xmath20 denotes the flavor , @xmath56 the quark mass , @xmath57 and @xmath58 with the fermi momentum @xmath59 . after some manipulation we easily find @xmath60 ^ 2 \right\ } , \label{fockene}\ ] ] where @xmath61 and @xmath62 , and we use the relation @xmath63 . equation ( [ fockene ] ) becomes @xmath64 for massless particles . for the interaction energy of nucleons ,
we use for simplicity the effective potential parametrized by densities @xcite , @xmath65 where , @xmath66 , @xmath67 , @xmath68 , and @xmath69 are adjustable parameters to satisfy the saturation properties of nuclear matter .
l50 mm we consider a system which consists of the hadron and quark phases .
we divide the whole space into equivalent and charge - neutral wigner - seitz cells with the size @xmath70 and the droplet size @xmath71 as illustrated in fig .
[ wsapp ] .
here we only consider the droplet phase .
following dft we begin with the thermodynamic potential in terms of particle density profiles and chemical potentials , @xmath72 and @xmath73 , @xmath74 to take into account the coulomb interaction , we write it in terms of the @xmath75-particle density profiles ; @xmath76 with @xmath77 being the particle charge ( @xmath78 for the electron ) , @xmath79 applying the laplacian @xmath80 on eq .
( [ coul ] ) , we can easily derive the poisson equation @xmath81 .
the coulomb interaction energy @xmath82 is then expressed as @xmath83 including the surface term , the total energy is expressed as @xmath84 where @xmath85 means the surface - energy density , @xmath86 boundary area , and @xmath87 and @xmath88 the energy densities in the quark and hadron phases . since we poorly know the details of the hadron quark interface , we simply approximate the surface energy as @xmath89 by using the surface tension @xmath90 .
each chemical potential is derived by the equation of motion @xmath91 , which reads @xmath92 , or @xmath93 as we have seen in the previous section , each phase may have a finite net charge . in the previous studies of smp
, the coulomb interaction between charged particles has been treated rather simply : the coulomb energy was added to the total energy by using the volume fraction @xmath20 and constant densities derived for two infinite matters @xcite .
thus the coulomb screening effect and rearrangement of charge densities are completely discarded .
one important point we would like to address here is , when there is the coulomb interaction , the gauge variance of chemical potentials should be taken into account .
differentiating the expression of chemical potential ( [ chempot ] ) with respect to the coulomb potential @xmath94 , we get the relation @xcite , @xmath95 where matrices @xmath96 and @xmath97 are defined as @xmath98 from these equations , the gauge invariance relation is derived , @xmath99 we can immediately see that the chemical potential is gauge variant from this equation : when the coulomb potential is shifted by a constant value , @xmath100 , the chemical potential should also be shifted @xmath101 to keep the gauge invariance .
note that the charge density should be observable and gauge invariant ; we shall see the constant shift of the coulomb potential is compensated by the redefinition of the chemical potential in the expression of the charge density .
the poisson equation is also gauge invariant because it is not changed by the constant shift of the coulomb potential .
therefore , if the poisson equation is not properly taken into account , the gauge invariance is obviously violated .
next , we present the explicit form of the thermodynamic potential for each particle . first , we consider the quark phase which consists of u , d and s quarks and electron .
according to the small current quark mass , we treat u and d quarks as massless particles and only s quark massive ( 150 mev ) in this phase .
electron is also treated as a massless particle .
next , we consider hadron phase which consists of proton , neutron and electron .
nucleons are treated as non - relativistic particles .
the total thermodynamic potential becomes @xmath102 here @xmath103 and @xmath104 are the thermodynamic potentials in the quark and hadron phases , and @xmath105 the surface energy .
the thermodynamic potentials in the quark phase are given by eqs .
( [ kinene ] ) , ( [ fockene ] ) , ( [ totene ] ) : @xmath106 , \\
\omega_{\mathrm{d } } & = & \int_0^r d^3r \left [ \frac{3\pi^{\frac{2}{3}}}{4 } \left ( 1 + \frac{2 \alpha_s}{3 \pi } \right ) \rho_{\mathrm{d}}^{\frac{4}{3}}(\bm{r } ) - \mu_{\mathrm{d } } \rho_{\mathrm{d}}(\bm{r } ) + \frac{1}{3}v_{\mathrm{coul}}(\bm{r } ) \rho_{\mathrm{d}}(\bm{r } ) \right ] , \\
\omega_{\mathrm{s } } & = & \int_0^r d^3r \left [ \epsilon_{\mathrm{s}}(\rho_{\mathrm{s}}(\bm{r } ) ) - \mu_{\mathrm{s } } \rho_{\mathrm{s}}(\bm{r } ) + \frac{1}{3}v_{\mathrm{coul}}(\bm{r } ) \rho_{\mathrm{s}}(\bm{r } ) \right ] , \\ \omega_\mathrm{em}^\mathrm{q } & = & \int_0^{r } d^3r \left [ -\frac{1}{8\pi e^2 } \left ( \nabla v_{\mathrm{coul}}(\bm{r } ) \right)^2 + \frac{\left ( 3 \pi^2 \rho_e(\bm{r } ) \right)^\frac{4}{3}}{4\pi^2 } - \mu_e \rho_e(\bm{r } ) + v_{\mathrm{coul}}(\bm{r } ) \rho_e(\bm{r } ) \right].\end{aligned}\ ] ] here , @xmath107 summarizes the electron and another coulomb contributions .
the energy density of s quark @xmath108 is a function of @xmath109 , which is explicitly expressed by using eqs .
( [ kinene ] ) and ( [ fockene ] ) as @xmath110 - \frac{\alpha_s}{\pi^3 } \left [ p_\mathrm{fs}^4(\bm{r } ) - \frac{3}{2 } m_\mathrm{s}^4 \left [ \frac{p_\mathrm{fs}(\bm{r})}{m_\mathrm{s } } \sqrt { 1 + \left ( \frac{p_\mathrm{fs}(\bm{r})}{m_\mathrm{s}}\right)^2 } \right.\right.\right.\nonumber\\ & & \left.\left.\left . - \ln \left ( \frac{p_\mathrm{fs}(\bm{r})}{m_\mathrm{s } } + \sqrt { 1 + \left ( \frac{p_\mathrm{fs}(\bm{r})}{m_\mathrm{s}}\right)^2 } \right ) \right]^2 \right ] \right\ } .\end{aligned}\ ] ] the thermodynamic potentials in the hadron phase read , @xmath111 \\
\omega_{\mathrm{n } } & = & \int_r^{r_w } d^3r \left [ \frac{3}{10m}\left ( 3 \pi^2 \right)^\frac{2}{3 } \rho^\frac{5}{3}_{\mathrm{n}}(\bm{r } ) - \mu_{\mathrm{n } } \left ( \rho_{\mathrm{p}}(\bm{r } ) , \rho_{\mathrm{n}}(\bm{r } ) \right ) \rho_{\mathrm{n}}(\bm{r } ) % \left ( \rho_{\mathrm{p}}(\bm{r } ) , \rho_{\mathrm{n}}(\bm{r } ) \right ) \right ] , \\
% \omega_{\mathrm{p } } & = & 2 \times \int d^3r \left [ \int_0^{p_{\mathrm{f } } } d^3p \sqrt{m_{\mathrm{p}}^2 + p^2 } - \left\ { g\left ( \rho_{\mathrm{p } } , \rho_{\mathrm{n } } \right ) - v_{\mathrm{coul } } \right\ } \rho_{\mathrm{p } } \right ] \\
\omega_{\mathrm{p } } & = & \int_r^{r_w } d^3r \left [ \frac{3}{10m}\left ( 3 \pi^2 \right)^\frac{2}{3 } \rho^\frac{5}{3}_{\mathrm{p}}(\bm{r } ) - \mu_{\mathrm{p}}\left ( \rho_{\mathrm{p}}(\bm{r } ) , \rho_{\mathrm{n}}(\bm{r } ) \right)\rho_\mathrm{p}(\bm{r } ) - v_{\mathrm{coul}}(\bm{r } ) \rho_{\mathrm{p}}(\bm{r } ) \right ] , \\
\omega_\mathrm{em}^\mathrm{h } & = & \int_r^{r_w } d^3r \left [ -\frac{1}{8\pi e^2 } \left ( \nabla v_{\mathrm{coul}}(\bm{r } ) \right)^2 + \frac{\left ( 3 \pi^2 \rho_e(\bm{r } ) \right)^\frac{4}{3}}{4\pi^2 } - \mu_e \rho_e(\bm{r } ) + v_{\mathrm{coul}}(\bm{r } ) \rho_e(\bm{r } ) \right].\end{aligned}\ ] ] here @xmath112 summarizes the electron and another coulomb contributions .
we get the expression of chemical potentials and the poisson equation from the equation of motion @xmath113 , where @xmath114 .
the poisson equation is explicitly written as @xmath115.\end{aligned}\ ] ] the chemical potentials for quarks in the quark phase are derived as @xmath116 + \frac{1}{3 } v_{\mathrm{coul}}(\bm{r } ) .\end{aligned}\ ] ] here , we use @xmath117 and @xmath118 . the nucleon chemical potentials in the hadron phase and the electron chemical potential
are @xmath119 note that , e.g. , the electron density profile @xmath120 is expressed as @xmath121 in a gauge invariant fashion ( cf .
( [ mc2 ] ) ) .
we impose the @xmath31 equilibrium in the quark and hadron phases and the chemical equilibrium at the quark - hadron boundary as in eqs .
( [ chemequ])-([chemeqp ] ) . the pressure contribution from the surface tension
is derived as @xmath122 with @xmath123 being the volume of the quark droplet . from eqs .
( [ omeq ] ) and ( [ omeh ] ) , the pressure of each phase is expressed as @xmath124 with @xmath125 being the volume of hadron phase .
therefore the pressure balance condition becomes @xmath126 which gives the droplet size @xmath71 .
finally the cell size @xmath70 is determined by the minimum condition for @xmath17 .
particle density profiles and @xmath71 , @xmath70 are completely determined for given @xmath8
. thus we can solve these coupled equations of motion consistently with gc .
note that the coulomb potential is included in a proper way and @xmath127 appears in almost all chemical potentials .
the coulomb potential is a functional of the charged - particle density profiles and in turn densities are functions of the coulomb potential . as a result ,
the poisson equation becomes highly non - linear .
since it is difficult to solve analytically , we solve it numerically without any approximation , whereas the linear approximation has been used in the previous work @xcite .
we show a case of quark droplet embedded in hadron matter for the volume fraction @xmath128 and the surface tension @xmath129 mev / fm@xmath1 .
we can see the coulomb screening effect in fig .
[ insta ] .
there is always a minimum in the thermodynamic potential with respect to the droplet radius @xmath71 without the coulomb screening , due to the balance between the coulomb energy and the surface energy .
however , once the coulomb screening is taken into account , the minimum disappears , which shows the mechanical instability caused by the coulomb screening effect .
if the surface tension is large enough , the geometrical structure ( droplet ) becomes mechanically unstable by the coulomb screening .
next , we show the density profiles .
we can see the flat density profile of particles in the left panel of fig .
[ densprof ] , where there is no coulomb screening effect . in the right panel , on the other hand , we can see the rearrangement of the charged particles as the coulomb screening effect .
let us consider the features of these density profiles in detail . in the quark phase , the densities of the negatively charged d , s
quarks and electron are reduced , while that of the positively charged u quarks are enhanced by the coulomb screening effect .
remember that the quark phase has a negative charge and the hadron phase a positive charge in the hadron - quark mixed phase .
r80 mm for the hadron phase , we can see the opposite effect by the coulomb screening : the electron number increases and the proton number is reduced .
we see that the charged particles in one phase are also affected by the charge of another phase : negatively charged particles in the quark phase are attracted to the surface , while positively charged particles ( u quarks ) are repelled from the surface . on the contrary ,
the positively charged particles in the hadron phase ( protons ) are attracted to the surface , while electrons are repelled .
although the rearrangement effect is not enough to establish the local charge neutrality , we can understand that the coulomb screening effect is to reduce the local charge in each phase . to draw the phase diagram of the hadron - quark mixed phase
, we have to minimize the thermodynamic potential with respect to the volume fraction , i.e. , we have to change the droplet radius @xmath71 and wigner - seitz cell size @xmath131 . without the coulomb screening effect ( left panel ) and with the coulomb screening effect ( right panel ) .
this is the case of @xmath132 mev and the volume fraction @xmath133 .
each density uniformly spreads in each phase and @xmath134 is constant ( = 0 ) in the left panel , while there is rearrangement of the charge densities and @xmath134 is spatially dependent in the right panel.,width=302 ] without the coulomb screening effect ( left panel ) and with the coulomb screening effect ( right panel ) .
this is the case of @xmath132 mev and the volume fraction @xmath133 .
each density uniformly spreads in each phase and @xmath134 is constant ( = 0 ) in the left panel , while there is rearrangement of the charge densities and @xmath134 is spatially dependent in the right panel.,width=302 ] we show the total thermodynamic potential for the droplet phase in fig .
[ omedef ] . comparing with uniform hadron matter and quark matter
, the droplet phase takes the smallest value of thermodynamic potential between a certain range of the baryon - number chemical potential . to clarify the difference between uniform matter and the droplet phase , we show the difference of thermodynamic potential between them in the right panel of fig . [ omedef ] .
we can see a large difference between our result ( @xmath135 mev / fm@xmath1 ) and that given by the bulk calculation in fig .
[ omedef ] .
( left panel ) and its difference from uniform hadron matter and quark matter @xmath136 ( right panel ) .
the dashed line `` bulk gibbs '' , in the right panel , shows the difference between the bulk calculation and uniform matter .
solid lines are the cases for the droplet phase with @xmath137 and 40 mev / fm@xmath1.,width=302 ] ( left panel ) and its difference from uniform hadron matter and quark matter @xmath136 ( right panel ) .
the dashed line `` bulk gibbs '' , in the right panel , shows the difference between the bulk calculation and uniform matter .
solid lines are the cases for the droplet phase with @xmath137 and 40 mev / fm@xmath1.,width=302 ] if the difference @xmath136 is negative , the mixed phase is an energetically favorable state . on the contrary ,
if @xmath136 is positive , the mixed phase is an unfavorable state . in the right panel of fig .
[ omedef ] , the point given by the maxwell construction is specified by a circle ( @xmath138 mev , @xmath139 ) , where the following relations are maintained , @xmath140 note that if we naively apply mc , we have to discard one of gc , @xmath141 in the absence of the coulomb potential .
the curve of `` bulk gibbs '' shows that the mixed phase is energetically favorable in the wide @xmath8 region , as is already seen in sec . 2 .
the difference @xmath136 for `` bulk gibbs '' is up to 10 mev/@xmath142 , while for the droplet phase ( @xmath135 mev / fm@xmath1 ) , it is only less than 2 mev/@xmath142 .
this shows that if we use the larger value of the surface tension , the mixed phase gets more unfavorable .
this feature is similar to that reported by heiselberg @xcite or alford @xcite .
we can see the region of the energetically favorable mixed phase in the right panel of fig .
[ omedef ] : the region becomes narrower than `` bulk gibbs '' , which means that the property of the mixed phase is closer to that given by mc .
thus we have seen that the mixed phase is very much affected by the coulomb screening effect and the surface effect , by strictly keeping gc . for the larger value of the surface tension , mc is effectively useful in the description of the hadron - quark mixed phase .
note that we never violate the condition for the charge chemical equilibrium @xmath143 to get these conclusions .
remember that the meaning of the condition in the presence of the coulomb potential is different from that in the absence of the coulomb potential .
in this note we have first seen how the bulk calculation is performed for the hadron - quark mixed phase by applying the gibbs conditions to the system consisting of two infinite matters .
it gives a wide density region for the mixed phase .
based on this result , some authors also suggested the structured mixed phase with various geometrical structures , by including the finite - size effects , the surface and coulomb energies @xcite .
we have examined the finite - size effects in the mixed phase by numerically solving the equations of motion for the particle densities and the poisson equation for the coulomb potential .
our framework is based on the idea of the density functional theory , which , we believe , is one of the best theories to treat the structured mixed phases .
we have demonstrated , by taking the droplet phase as an example , that the coulomb screening effect and rearrangement of the charge densities play an important role for the mechanical instability as well as the energy of the mixed phase . as a result
we have seen that the region of the mixed phase is highly restricted by the coulomb screening effect as well as the surface energy .
we have also seen that eos gives the similar behavior to that given by the maxwell construction , whereas the maxwell construction is apparently incorrect in a system with more than one chemical potential : the maxwell construction is effectively useful in the description of the mixed phase , even in this case . as another case of more than one chemical potential ,
kaon condensation has been also studied @xcite and the result is similar to the present case .
we have included the surface tension , but its definite value is not clear and many authors treated it as a free parameter .
there are also many estimations for the surface tension at the hadron - quark interface in lattice qcd @xcite , in shell - model calculations @xcite and in model calculations based on the dual - ginzburg landau theory @xcite .
if we have the realistic value of the surface tension , we can reasonably bring out smp in the hadron - quark phase transition . as these system corresponds to neutron star matter
, we have seen that the mixed phase should be narrow by the finite - size effects .
our result would restrict the allowed smp region in neutron stars which is suggested by glendenning @xcite .
it could be said that they should change the property of neutron stars especially in equation of state @xcite .
we have assumed in relation to phenomenological implications here that temperature is zero .
it would be much interesting to include the finite - temperature effect .
then it is possible to draw the phase diagram in the @xmath144 plane and we can study the properties of the deconfinement phase transition . in this study
we have used a simple model for quark matter to figure out the finite - size effects in the structured mixed phase .
however , it has been suggested that the color superconductivity is a ground state of quark matter @xcite . to get more realistic picture of the hadron - quark phase transition
, we will need to take into account color superconductivity .
in the recent studies the mixed phase has been also studied @xcite .
we thank to d. n. voskresensky and t. tanigawa for their fruitful discussion . t.e . and t.t .
acknowledge the support and kind hospitality of the joint institute for nuclear research ( jinr ) when we stayed in dubna .
m. alford , k. rajagopal and f. wilczek , nucl .
b * 64 * ( 1999 ) 443 , d. bailin and a. love , phys . rep . * 107 * ( 1984 ) 325 . as resent reviews , m. alford , hep - ph/0102047 ; k. rajagopal and f. wilczek , hep - ph/0011333 , d. h. rischke , prog . part .
* 52 * ( 2004 ) 197 d. n. voskresensky , m. yasuhira and t. tatsumi , phys
b541 * ( 2002 ) 93 ; nucl . phys .
* a723 * ( 2003 ) 291 ; t. tatsumi , m. yasuhira and d. n. voskresensky , nucl .
a718 * ( 2003 ) 359c ; t. tatsumi and d. n. voskresensky , nucl - th/0312114 . | a structured mixed phase consisting of quark and hadron phases is numerically studied with the coulomb screening effect and the surface effect .
we carefully introduced the coulomb potential , so that a geometrical structure becomes mechanically unstable when the surface tension is large .
charge densities are largely rearranged by the screening effect , and thereby the equation of state shows the similar behavior to that given by the maxwell construction .
therefore , although bulk calculations with the gibbs conditions show that the mixed phase may exist in a wide density region , we can see it is restricted to a narrow density region by the surface effect and the coulomb screening effect .
* hadron - quark matter phase transition in neutron stars * tomoki endo@xmath0 , toshiki maruyama@xmath1 , satoshi chiba@xmath1 and toshitaka tatsumi@xmath2 @xmath2 department of physics , kyoto university , kyoto 606 - 8502 , japan + @xmath1 advanced science research center , japan atomic energy research institute , tokai , ibaraki 319 - 1195 , japan + |
it is widely accepted that quantum entanglement is an essential ingredient for the implementation of quantum information processing devices @xcite .
creating highly entangled multi - particle states is therefore one of the most challenging goals of modern experimental quantum mechanics .
quantum entanglement of two or more systems leads to correlations between observables of the systems that can not be explained on the basis of local realistic theories .
quantum entanglement lies at the heart of the difference between the quantum and the classical world . due to its vast application
, quantum entanglement has been studied in different systems such as optomechanical systems@xcite , bose einstein condensates ( bec ) trapped in double quantum well @xciteand in optical lattices @xcite .
entanglement has also been observed experimentally at nist group @xcite , where entanglement state of four - ions was created successfully by using the scheme proposed by sorensen and molmer @xcite . in order to build a quantum information network using atomic systems ,
quantum state exchange between light and matter is an essential ingredient .
one prefers matter over photons as quantum memory elements since experimentally it is difficult to localize and store photons .
stimulated raman adiabatic passage is considered to be a very convenient technique of storing light and has been used in single atom cavity quantum electrodynamics for transfer of qubits between atoms and photons and for building quantum logic gates @xcite . mapping quantum states to collective atomic spin systems @xcite and quantized vibrational states of trapped atoms @xcite by means of stimulated raman
adiabatic passage technology and electromagnetically induced transparency @xcite has been discussed earlier .
new possibilities for cavity opto - mechanics may emerge by combining the tools of cavity quantum electrodynamics(qed ) with those of ultracold gases @xcite . placing an ensemble of atoms inside a high - finesse cavity enhances the atom - light interaction because the atoms collectively couple to the same light mode .
the motional degrees of freedom of ultracold atomic gases represent a new source of long - lived coherence affecting light - atom interaction .
nonlinear optics arising from this long - lived coherent motion of ultracold atoms trapped within a high - finesse fabry - perot cavity was reported recently @xcite .
strong optical nonlinearities were observed even at low mean photon number of @xmath0 .
this nonlinearity also gives rise to bistability in the transmitted probe light through the cavity .
experimental implementation of a combination of cold atoms and cavity qed ( quantum electrodynamics ) has made significant progress @xcite .
theoretically there have been some interesting work on the correlated atom - field dynamics in a cavity .
it has been shown that the strong coupling of the condensed atoms to the cavity mode changes the resonance frequency of the cavity @xcite .
finite cavity response times lead to damping of the coupled atom - field excitations @xcite .
the driving field in the cavity can significantly enhance the localization and the cooling properties of the system@xcite .
it has been shown that in a cavity the atomic back action on the field introduces atom - field entanglement which modifies the associated quantum phase transition @xcite . the light field and
the atoms become strongly entangled if the latter are in a superfluid state , in which case the photon statistics typically exhibits complicated multimodal structures @xcite .
a coherent control over the superfluid properties of the bec can also be achieved with the cavity and pump @xcite .
recently , a new approach was proposed which is based on all optical measurements that conserve the number of atoms .
it was shown that atomic quantum statistics can be mapped on transmission spectra of high - q cavities , where atoms create a quantum refractive index .
this was shown to be useful for studying phase transitions between mott insulator and superfluid states since various phases show qualitatively distinct spectra @xcite .
motivated by such interesting developments in the field of cavity electrodynamics of bose einstein condensates , we propose in this study a new scheme of entangling two bose - einstein condensates ( bec ) in a double cavity set - up by transferring the entanglement of quantum - correlated light fields produced from a nondegenerate parametric amplifier ( nopa ) to the collective density excitations of the two becs .
in this section , we show that the quantum state of the input light field can be transferred to a bec trapped in a cavity .
we consider an elongated cigar shaped bose - einstein condensate(bec ) of @xmath1 two - level @xmath2 atoms in the @xmath3 state with mass @xmath4 and frequency @xmath5 of the @xmath6 transition of the @xmath7 line of @xmath2 , strongly interacting with a quantized single standing wave cavity mode of frequency @xmath8 as shown in fig .
1 . the standing wave that forms in the cavity results in an one - dimensional optical lattice potential .
the cavity field is also coupled to external fields incident from one side mirror .
it is well known that high - q optical cavities can significantly isolate the system from its environment , thus strongly reducing decoherence and ensuring that the light field remains quantum - mechanical for the duration of the experiment .
we also assume that the induced resonance frequency shift of the cavity is much smaller than the longitudinal mode spacing , so that we restrict the model to a single longitudinal mode . in order to create an elongated bec
, the frequency of the harmonic trap along the transverse direction should be much larger than one in the axial ( along the direction of the optical lattice ) direction .
the system is also coherently driven by a laser field with frequency @xmath9 through the cavity mirror . here the two mirrors @xmath10 and @xmath11 are fixed .
this system is modelled by a jaynes - cummings type hamiltonian ( @xmath12 ) in the rotating and dipole approximation .
is incident from the left mirror and from the same mirror the reflected light from the bec is taken out as @xmath13 .
faraday oscillator ( @xmath14 ) enables a unidirectional coupling . ]
@xmath15 + i \hbar \sqrt{2 \gamma}(c_{in}c^{\dagger}-c_{in } ^ { * } c ) , \ ] ] where @xmath16 and @xmath17 are the pauli matrices and @xmath18 and @xmath19 are the large atom - pump and cavity - pump detuning , respectively . .
@xmath20 is the space dependent atom field coupling . @xmath21 and @xmath22 are the annihilation and creation operators of the intracavity field .
the input laser field populates the intracavity mode which couples to the atoms through the dipole interaction .
the field in turn is modified by the back - action of the atoms .
the system is open as the cavity field is damped by the photon leakage through the coupling mirror .
we are considering a system with large detuning and hence spontaneous emission is ignored and hence we can adiabatically eliminate the excited state using heisenberg equation of motion @xmath23 $ ] , which yields the single particle hamiltonian @xmath24\ ] ] the parameter @xmath25 is the optical lattice barrier height per photon and represents the atomic backaction on the field .
also @xmath26 is the external classical potential . here
we will always take @xmath27 . in this case
the condensate is attracted to the nodes of the light field and hence the lowest bound state is localized at these positions which leads to a reduced coupling of the condensate to the cavity compared to that for @xmath28 . along @xmath29 ,
the cavity field forms an optical lattice potential of period @xmath30 and depth ( @xmath31 ) .
we now write the hamiltonian in a second quantized form including the two body interaction term .
@xmath32 where @xmath33 is the field operator for the atoms .
here @xmath34 is the two body @xmath35-wave scattering length .
the corresponding opto - mechanical - bose - hubbard ( ombh ) hamiltonian can be derived by writing @xmath36 , where @xmath37 is the wannier function and @xmath38 is the corresponding annihilation operator for the bosonic atom at the @xmath39 site . retaining only the lowest band with nearest neighbor interaction , we have @xmath40 j_{o } \sum_{i } b_{i}^{\dagger } b_{i}+\frac{u}{2 } \sum_{i } b_{i}^{\dagger}b_{i}^{\dagger } b_{i } b_{i}\ ] ] where @xmath41 here , we have neglected the tunneling terms @xmath42 and @xmath43 , as they can be made small by varying the optical lattice depth . under this approximation ,
the matter - wave dynamics is not essential for light scattering . in experiments ,
such a situation can be realized because the time scale of light measurements can be much faster than the time scale of atomic tunneling .
the dynamics of the system can be described by the following quantum langevin equations @xmath44 @xmath45j_{o}b_{i}-i\frac{u}{\hbar}b^{\dagger}_{i } b_{i } b_{i}-\gamma b_{i}+\xi.\ ] ] here @xmath46 is the cavity decay rate , @xmath47 is the condensate decay rate and @xmath48 is the classical thermal noise operator of the bec and has the following correlation function at temperature @xmath49 .
@xmath50=@xmath51=@xmath52,@xmath53 and @xmath54 . the noise operators for the input field obey the following correlation functions : @xmath55 , @xmath56.the thermal noise input for the bec is provided by the thermal cloud of atoms .
we now write each canonical operator of the system as a sum of its steady state mean value and a small fluctuation with zero mean value i.e @xmath57 and @xmath58 and linearize to obtain the following heisenberg - langevin equations for the fluctuation operators .
@xmath59 @xmath60b -i u_{eff } b^{\dagger}-i g_{c}(c^{\dagger}c_{s}+c_{s}^{\dagger}c)-\gamma b+ \xi,\ ] ] here , @xmath61 , @xmath62 , @xmath63 , @xmath64 is the detuning with respect to the renormalized resonance .
@xmath1 is the total number of atoms in @xmath65 sites .
as before , we assume negligible tunneling ( @xmath66 ) and hence we drop the site index @xmath67 from the atomic operators .
we now make the transformations @xmath68 , @xmath69 and @xmath70 . neglecting the fast rotating terms ,
we get the following equations @xmath71 @xmath72 if we assume that decay rate @xmath46 of the cavity field is very large so that we can adiabatically eliminate the dynamics of the cavity mode . consequently we have the steady state value of @xmath73 . @xmath74 on substituting [ steady_1 ] in [ fluc_2 ] and assuming @xmath75 for simplicity we get @xmath76 here @xmath77 .
if @xmath78 , the statistics of the input light field can be transferred to the collective density excitations of the bec .
here we now utilize the results obtained in the previous section to show that how we can transfer entanglement of a pair of quantum correlated light fields into a pair of becs in two independent cavity which are physically separated .
the scheme is shown in figure 2 .
the quantum correlated light fields are generated by a nopa .
the output fields are frequency degenerate but polarization nondegenerate .
the coupling between the two intracavity field modes @xmath79 and @xmath80 can be represented by @xmath81 . here
@xmath82 is the coupling strength that is proportional to the nonlinear susceptibility of the intracavity medium and the intensity of the coherent pump field .
we take @xmath83 as the damping rate of the cavity modes .
the equations of motion for the mode operators can be written as @xmath84 @xmath85 where @xmath86 and @xmath87 are the vacuum input fields of the two cavity modes of nopa .
the output fields from the nopa follows from the boundary conditions @xmath88 @xmath89 the quadrature amplitudes @xmath90 , @xmath91 , @xmath92 in fourier space are found as : @xmath93\ ] ] @xmath94\ ] ] the highly correlated light fields from the nopa are incident on the two space separated cavities with two identical becs .
we can now write the bec mode operators of the two condensates as @xmath95 @xmath96 here we assume that the coupling between the nopa and the cavities is unidirectional . if system parameters are chosen such that @xmath97 and @xmath98 can be regarded as quantum white noise operators , then the variances of the positions ( @xmath99 ) and momenta ( @xmath100 ) of the two becs are given by @xmath101 } \right\}+\frac { \gamma ( 1 + 2 n_{t})}{b}\ ] ] as a function of @xmath102 for two values of the detuning @xmath103 ( thin line ) and @xmath104 ( thick line ) .
the other parameters are @xmath105 , @xmath106 , @xmath107 , @xmath108 and @xmath109 .
quantum entanglement is said to exist when @xmath110 . ] here , @xmath111 , @xmath112 , @xmath113 and @xmath114 .
the two identical becs are entangled if @xmath110 . for certain set of parameters , the variances @xmath115 @xmath116 @xmath117 @xmath118 @xmath119 i.e an epr state in the position and momenta of the two becs . as evident from eqn .
[ variance_1 ] , the thermal contribution destroys the entanglement . a plot of epr variance @xmath120 as a function of @xmath102 for two values of the detuning @xmath103 ( thin line ) and @xmath104 ( thick line )
is shown in fig .
quantum entanglement is said exist when @xmath110 .
clearly we see that significant entanglement is achieved when @xmath121 .
further we observe that for a large detuning @xmath122 , the entanglement is more but the range of @xmath123 where this entanglement can be achieved is reduced . on the other hand for a smaller detuning , the entanglement is reduced but the range over which entanglement is achieved is enhanced .
an epr state can thus be reached for the collective density excitations of the two becs by keeping @xmath124 and @xmath125 . to demonstrate that the dynamics investigated here are within experimental reach
, we discuss the experimental parameters from @xcite : a bec of typically @xmath126 @xmath127 atoms is coupled to the light field of an optical ultra high - finesse fabry - perot cavity .
the atom - field coupling @xmath128 @xcite ( @xmath129 @xcite ) is greater than the decay rate of the intracavity field @xmath130 @xcite ( @xmath131 @xcite ) .
typically atom - pump detuning is @xmath132 .
the rate @xmath133 at which atoms are coupled out of the bec is about @xmath134 @xcite .
the kinetic energy and potential energy contribution @xmath135 is about @xmath136 @xcite(@xmath137 @xcite).the energy of the cavity mode decreases due to the photon loss through the cavity mirrors , which leads to a reduced atom - field coupling .
photon loss can be minimized by using high - q cavities .
our proposed scheme relies crucially on the fact that coherent dynamics dominate over the losses .
it is important that the characteristic time - scales of coherent dynamics are significantly faster than those associated with losses .
in conclusion , we have proposed a novel scheme to transfer the quantum entanglement between two light fields emerging from a nondegenerate parametric amplifier to the collective density excitations of two bose einstein condensates which are physically separated in two independent optical cavities .
an epr state of the collective density excitations can be created by a judicious choice of the parameters .
this study could be of use in studying entanglement in macroscopic quantum objects and for high precision metrology .
99 r. f. werner quantum information processing(springer tracts in modern physics ) * 173 * , ( 2001 ) .
s. mancini et al . , phys .
lett . * 88 * , 12 ( 2002 ) ; m. pinard et al .
, europhys . lett . * 72 * 747 ( 2005 ) ; j. zhang et al .
a * 68 * 013808 ( 2003 ) .
j. chen et al .
a * 360 * 429 2007 . l. you phys .
letts * 90 * 030402 2003 . c. a. sackett et al .
, naturelondon * 404 * , 256 ( 2000 ) .
a. srensen and k. mlmer , phys .
. lett . * 82 * , 1971 ( 1999 ) ; k. mlmer and a. srensen , phys .
lett . * 82 * , 1835 ( 1999 ) ; j. mod . opt . *
47 * , 2515 ( 2000 ) ; a. srensen and k. mlmer , phys .
a * 62 * , 022311 ( 2000 ) .
a. s .parkins et al . , phys . rev
. letts . * 71 * 3095 ( 1993 ) ; t. pellizzari et al .
rev . letts . * 75 * 3788 ( 1995 ) ; j. i. cirac et al .
letts . * 78 * 3221 ( 1997 ) .
a. e. kozhekin et al . , phys .
a * 62 * , 033809 ( 2000 ) .
parkin and h. j. kimble , j. opt .
b : quantum semiclassical opt . , * 1 * , 496 ( 1999 ) .
m. d. lukin et al .
letts . , textbf84 , 4232 ( 2000 ) .
a. bhattacherjee , phys .
a , * 80 * , 043607 ( 2009 ) . f. brennecke et al . , science * 322 * , 235 ( 2008 ) .
k. w. murch et al . ,
nature physics * 4 * , 561 ( 2008 ) .
b. nagorny , tn .
elssser and a. hemmerich , phys .
letts . , * 91 * , 153003 ( 2003 ) .
j. a. sauer , k.m .
fortier , m.s.chang , c.d .
hamley and m.s .
chapman , phys .
a , * 69 * , 051804(r ) ( 2004 ) .
anton ttl , stephan ritter , michael khl and tilman esslinger , phys .
letts . , * 95 * , 090404 ( 2005 ) .
p. horak , s. m. barnett and h. ritsch , phys .
a , * 61 * , 033609 ( 2000 ) .
p. horak , h. ritsch , phys .
a * 63 * , 023603 ( 2001 ) .
a. griessner , d. jaksch and p. zoller , j. phys .
b,*37 * , 1419 ( 2004 ) . c. maschler and h. ritsch , opt
. comm . * 243 * , 145 ( 2004 ) . c. maschler and h. ritsch , phys .
* 95 * , 260401 ( 2005 ) .
w. chen , d. meiser and p. meystre , phys .
a , * 75 * , 023812 ( 2007 ) .
a. bhattacherjee , optics communications , * 281 * , 3004 ( 2008 ) .
i. b. mekhov , c. maschler and h. ritsch , nature physics , * 3 * , 319 ( 2007 ) . | we propose a scheme to transfer the quantum state of light fields to the collective density excitations of a bose einstein condensate ( bec ) in a cavity .
this scheme allows to entangle two becs in a double cavity setup by transferring the quantum entanglement of two light fields produced from a nondegenerate parametric amplifier ( nopa ) to the collective density excitations of the two becs .
an epr state of the collective density excitations can be created by a judicious choice of the system parameters . |
here we briefly outline basic steps and main results of the spin - wave calculations for the energy spectrum and the magnon relaxation rates of the @xmath143@xmath144 antiferromagnet on a honeycomb lattice .
the harmonic spin - wave analysis of the nearest - neighbor heisenberg honeycomb - lattice antiferromagnet can be found , for example , in @xcite .
geometry of exchange bonds of the considered model is schematically shown in fig .
[ suppl : lattice ] .
the unit cell of the antiferromagnetic structure coincides with the crystal unit cell and contains two oppositely aligned spins @xmath145 and @xmath146 in positions @xmath147 and @xmath148 .
the elementary translation vectors are defined as @xmath149 and @xmath150 .
the lattice constant in bani@xmath3(po@xmath4)@xmath3 is equal to @xmath151 .
the reciprocal lattice basis is @xmath152 and @xmath153 .
the volume of the brilouin zone is @xmath154 .
@xmath144 model in a honeycomb lattice .
] the spin hamiltonian includes heisenberg exchange interactions between first- and third - neighbor spins together with the single - ion anisotropy : @xmath155 \ . \nonumber\end{aligned}\ ] ] here @xmath156 denotes spin in the unit cell @xmath157 and so on .
the microscopic parameters for bani@xmath3(po@xmath4)@xmath3 ( @xmath158 ) were determined from the magnon dispersion as @xmath159 mev , @xmath160 mev , and @xmath161 mev @xcite .
the second - neighbor exchange was estimated to be much smaller @xmath162 mev and is neglected in the following .
applying the holstein - primakoff transformation for two antiferromagnetic sublattices and performing the fourier transformation @xmath163 we obtain the harmonic part of the boson hamiltonian @xmath164 \ , \nonumber\end{aligned}\ ] ] where we use the shorthand notations @xmath165 with @xmath166 and @xmath167 diagonalization of the quadratic form ( [ suppl : h2 ] ) with the help of the canonical bogolyubov transformation yields @xmath168 \ , \ ] ] where excitation energies are @xmath169 & & \varepsilon_\beta({\bf k } ) = s \sqrt { ( 3j_{13 } + |f_{\bf k}| ) ( 3j_{13 } - |f_{\bf k}| + 2d ) } \ .\end{aligned}\ ] ] the first magnon branch is gapless , @xmath170 , and reaches the maximum value of @xmath171 at @xmath172 $ ] with @xmath173 in the reciprocal lattice units .
the second branch describes optical magnons with a finite energy gap at @xmath174 @xmath175 the maximum of the optical branch @xmath176 is close to ( [ suppl : omax ] ) .
in the long - wavelength limit @xmath177 the energy of the acoustic branch has linear dispersion @xmath178 with the spin - wave velocity @xmath179 for the optical branch one finds @xmath180 with @xmath181 mev@xmath88 for bani@xmath3(po@xmath4)@xmath3 . for small @xmath182 the bogolyubov transformation
can be written explicitly in the following way .
first , we transform from the original holstein - primakoff bosons @xmath183 and @xmath184 to their linear combinations : @xmath185 the fourier transformed hamiltonian ( [ suppl : h2 ] ) takes the following form @xmath186 \ , .
\nonumber\end{aligned}\ ] ] second , the standard @xmath187@xmath188 transformation is applied separately for @xmath189 and @xmath190 bosons . in particular , for the acoustic branch , @xmath191 , we obtain @xmath192 where @xmath193 and @xmath194 . in the case of bani@xmath3(po@xmath4)@xmath3 the two dimensionless constants
are @xmath195 and @xmath196 .
similarly , for optical magnons with @xmath197 we obtain @xmath198 with @xmath199
for a collinear antiferromagnet the interaction between spin - waves is described by four - magnon terms in the bosonic hamiltonian .
the four - magnon terms of the exchange origin are expressed as @xmath200 where @xmath201 stands for @xmath202 etc .
the single - ion anisotropy contributes @xmath203 performing transformation from @xmath204 , @xmath205 to @xmath206 , @xmath207 we obtain various magnon - magnon terms .
the scattering of optical ( @xmath16 ) magnons on each other , which will be referred to as the roton - roton interaction , can be straightforwardly written as @xmath208 derivation of the roton - phonon interaction ( scattering of the optical magnon on the acoustic one , @xmath16 on @xmath15 ) is more involved and we obtain an estimate as @xmath209 the individual terms in the magnon - magnon interaction obtained from ( [ hj ] ) and ( [ hd ] ) applying the bogolyubov transformation are proportional to @xmath210 and diverge for scattering processes involving acoustic magnons , see ( [ suppl : uva ] )
. however , the leading @xmath211 and the subleading singularity @xmath212 cancel out in their net contribution and @xmath213 in agreement with the hydrodynamic approach @xcite . local modulation of magnetic coupling constants due to structural disorder , etc .
, will result in _ independent _ variations of @xmath134- and @xmath133-terms in magnon - magnon interaction in ( [ hj ] ) and ( [ hd ] ) .
thus , the resultant impurity - assisted magnon - magnon interaction will retain the same structure as the magnon - magnon interaction , with two important differences .
first , the momentum in such a scattering is not conserved , and , second , the variation of @xmath134 ( @xmath21 ) is associated only with ( [ hj ] ) and the variation @xmath22 will contain only ( [ hd ] ) part .
since such variations are independent , it suffices to consider one of them and treat the associated constant as a free parameter .
the most important consequence of this consideration is that , in the impurity scattering , there is no cancellation of the individual terms that are proportional to @xmath214 , compared to the case of magnon - magnon scattering in ( [ vrp ] ) discussed above where such a cancellation does take place .
thus , in the long - wavelength limit , @xmath51 , with a coefficient proportional to the impurity concentration and strength of the disorder .
the lowest - order diagram for the magnon self - energy calculated using matsubara technique is @xmath215 \nonumber\end{aligned}\ ] ] with @xmath216 . then the damping rate is @xmath217 \nonumber\\ & \times&\delta(\varepsilon_{\bf k } + \varepsilon_{\bf q } -\varepsilon_{\bf q ' } - \varepsilon_{\bf k'})\ , .
\label{gammak}\end{aligned}\ ] ] first , we consider the roton - roton scattering processes .
the low - temperature asymptote of ( [ gammak ] ) in this case is obtained by taking @xmath218 and keeping the leading exponentially small occupation factor .
then , for an optical magnon with @xmath174 in two dimensions @xmath219 performing integration in ( [ gamma01 ] ) and using parameters for bani@xmath3(po@xmath4)@xmath3 discussed above we obtain @xmath220 } \ . \label{gamma02rr}\ ] ] without going into details , there exist another channel of scattering that corresponds to a conversion of two rotons into two high - energy phonons , @xmath221 , which leads to the decay rate of the optical mode of the same exponential form as in ( [ gamma02rr ] ) with a numerical coefficient of the same order . | we demonstrate that local modulations of magnetic couplings have a profound effect on the temperature dependence of the relaxation rate of optical magnons in a wide class of antiferromagnets in which gapped excitations coexist with acoustic spin waves . in a two - dimensional collinear antiferromagnet with an easy - plane anisotropy ,
the disorder - induced relaxation rate of the gapped mode , @xmath0 , greatly exceeds the magnon - magnon damping , @xmath1 , negligible at low temperatures .
we measure the lifetime of gapped magnons in a prototype @xmath2 antiferromagnet bani@xmath3(po@xmath4)@xmath3 using a high - resolution neutron - resonance spin - echo technique and find experimental data in close accord with the theoretical prediction .
similarly strong effects of disorder in the three - dimensional case and in noncollinear antiferromagnets are discussed . _
introduction._the recent development of the neutron - resonance spin - echo technique has led to dramatic improvement of the energy resolution in neutron - scattering experiments @xcite . when applied to elementary excitations in magnetic insulators
, this technique allows one to measure magnon linewidth with the @xmath5ev accuracy compared to the mev resolution of a typical triple - axis spectrometer .
damping of quasiparticles depends fundamentally on the strength of their interactions with each other and with impurities , information not accessible directly by other measurements .
although theoretical studies of magnon damping in antiferromagnets ( afs ) go back to the 1970s @xcite , a comprehensive comparison between theory and experiment is still missing , mainly due to the lack of experimental data .
magnon - magnon scattering is traditionally viewed as the leading source of temperature - dependent magnon relaxation rates in afs @xcite .
another common relaxation mechanism in solids is the lattice disorder , which is responsible for a variety of the low - temperature effects , such as residual resistivity of metals @xcite and finite linewidth of antiferromagnetic resonances @xcite .
however , _ temperature - dependent _ effects of disorder are usually neglected because of the higher powers of @xmath6 in impurity - induced relaxation rates compared to leading scattering mechanisms and of the presumed dilute concentration and weakness of disorder .
the closest analogy is the resistivity of metals , in which the @xmath7 term is due to lattice imperfections and the temperature - dependent part is due to quasiparticle scattering . in this work ,
we demonstrate that scattering on the spatial modulations of magnetic couplings should completely dominate the low - temperature relaxation rate of gapped excitations in a wide class of afs .
such modulations , produced by random lattice distortions , yield scattering potential for propagating magnons and , at the same time , modify locally their interactions . for an illustration , we consider an example of the two - dimensional ( 2d ) easy - plane af with one acoustic and one gapped excitation branch .
in addition to potential scattering , responsible for a finite damping @xmath8 of optical magnons , see fig .
[ fig : diagrams](a ) , there exists an impurity - assisted temperature - dependent scattering of gapped magnons on thermally - excited acoustic spin waves , see fig .
[ fig : diagrams](c ) , which yields @xmath9 . despite the presumed smallness of impurity concentration @xmath10 , at low temperatures this mechanism dominates over the conventional magnon - magnon scattering , fig .
[ fig : diagrams](b ) , which carries a much higher power of temperature : @xmath11 .
we have performed resonant neutron spin - echo measurements with a few @xmath5ev resolution on a high - quality sample of bani@xmath3(po@xmath4)@xmath3 , a prototype 2d planar af @xcite .
we find that the theory describes very well the experimental data for the linewidth of optical magnons .
similar dominance of the impurity - assisted magnon - magnon scattering should persist in the 3d afs and is even more pronounced in the noncollinear afs .
we propose further experimental tests of this mechanism .
0.9 cm ) .,title="fig : " ] _ theory._we begin with the spin hamiltonian of a collinear af with an easy - plane anisotropy induced by the single - ion term @xmath12 : @xmath13 two examples are the nearest - neighbor afs on square and honeycomb lattices .
the latter model , with the non - frustrating third - neighbor exchange , is relevant to the spin-1 antiferromagnet @xmath14 @xcite discussed below . as a consequence of broken @xmath2 symmetry , excitation spectrum in the ordered antiferromagnetic state possesses acoustic ( @xmath15 ) and gapped ( @xmath16 ) magnon branches : @xmath17 see fig .
[ fig : diagrams](d ) for a sketch .
explicit expressions for @xmath18 , @xmath19 , and @xmath20 for bani@xmath3(po@xmath4)@xmath3 are provided in @xcite .
defects are present in all crystals .
while vacancies and substitutions may be eliminated in some materials , inhomogeneous lattice distortions remain an intrinsic source of disorder , inducing weak random variations @xmath21 and @xmath22 of microscopic parameters in the spin hamiltonian ( [ h ] ) @xcite .
both types of randomness have qualitatively the same effect on magnon lifetimes .
for example , local modification of the single - ion anisotropy @xmath23 generates scattering potential for magnons @xmath24 where @xmath25 , @xmath26 , and @xmath27 are the bogolyubov transformation parameters . for optical magnons at @xmath28 ,
the momentum dependence is not important , @xmath29 . for bond disorder ,
all expressions are the same with a substitution @xmath30 and an additional phase factor , which depends on bond orientation and disappears after impurity averaging . for the gapped magnons with @xmath31 , scattering amplitude in the second born approximation , fig .
[ fig : diagrams](a ) , averaged over spatial distribution of impurities is @xcite @xmath32 where @xmath10 is the impurity concentration , @xmath33 is the averaged impurity potential , and @xmath34 is the magnon bandwidth @xcite .
thus , in 2d , conventional impurity scattering results in a finite zero - temperature relaxation rate of the gapped magnons . at low temperatures , the principal scattering channel for optical magnons is due to collisions with the thermally excited acoustic spin waves with @xmath35 .
all other processes are either forbidden kinematically or exponentially suppressed . in this case
we can consider only @xmath36 terms in the magnon - magnon interaction : @xmath37 where the first and the second row correspond to the conventional and to the impurity - assisted magnon - magnon scattering , respectively , with @xmath38 .
the latter is of the _ same _ origin as the conventional impurity scattering in ( [ himp1 ] ) since @xmath22 and @xmath21 also modify locally interactions among magnons @xcite . in the one - loop approximation , ( [ hmm ] ) and ( [ 1h4imp ] )
yield the self - energies of figs .
[ fig : diagrams](b ) and ( c ) .
applying standard matsubara technique , relaxation rates can be expressed as @xmath39 where @xmath40 , @xmath41 , and @xmath42 is the bose factor .
there are two important differences between @xmath43 and @xmath44 in ( [ 1gmm ] ) and ( [ 1gimp ] ) .
first , the total momentum is not conserved for impurity scattering .
this relaxes kinematic constraints of the 4-magnon scattering processes , but requires instead integration over the extra independent momentum @xmath45 .
second and most crucial , interaction vertices @xmath46 and @xmath47 show very different long - wavelength behavior as @xmath48 , @xmath49 .
we calculate them using the approach similar to @xcite , and find that in the long - wavelength limit magnon - magnon interaction ( [ hmm ] ) is @xmath50 , in accordance with the hydrodynamic limit @xcite .
however , for the impurity - assisted scattering ( [ 1h4imp ] ) , interaction is @xmath51 .
this can be understood as a consequence of an effective long - range potential for acoustic magnons produced by the gaped magnon while in the vicinity of an impurity .
the leading @xmath6-dependence of @xmath52 and @xmath53 can be calculated now using ( [ wk ] ) and approximating interaction vertices with their long - wavelength expressions .
the main contribution to the integrals in ( [ 1gmm ] ) and ( [ 1gimp ] ) is determined by acoustic magnons with @xmath54 .
then , a straightforward power counting yields @xmath55 where @xmath56 @xcite .
thus , the inverse lifetime of an optical magnon is proportional to @xmath57 in 2d .
a generalization to higher dimensions gives @xmath58 .
the @xmath59-law for the relaxation rate of optical magnons in 3d afs was previously predicted in @xcite .
we note that for a given model , the effect of magnon - magnon scattering in ( [ 1gmm_est1 ] ) can be calculated using microscopic parameters , thus putting strict bounds on its magnitude . the same calculation for @xmath60 proceeds via the following integral : @xmath61 where @xmath62 with @xmath63 , @xmath64 , and we used the relation between @xmath65 in ( [ 1h4imp ] ) and @xmath66 in ( [ himp1 ] ) .
the nave power counting in ( [ 1gimp_est ] ) already gives @xmath67 , while a more careful consideration shows further enhancement of the scattering as the integrals formally diverge [ logarithmically ] in the @xmath68 region , demonstrating an important role of the long - wavelength magnons in 2d .
this divergence is similar to the one in the problem of finite @xmath69 ordering temperature in 2d and is regularized similarly by introducing low - energy cutoff .
the cutoff is either due to a 3d - crossover as in the case of some cuprates @xcite , or a weak in - plane anisotropy that induces small gap @xmath70 in the acoustic branch , the case directly relevant to the current work @xcite .
combining ( [ tau_0 ] ) and ( [ 1gimp_est ] ) we obtain impurity - induced relaxation rate of gapped magnons @xmath71 \ , , \end{aligned}\ ] ] where both @xmath72 and @xmath73 are proportional to @xmath10 and to the average strength of disorder @xmath74 . as a result , the impurity scattering leads to a relaxation rate that carries a significantly lower power of temperature than the magnon - magnon scattering mechanism . therefore , despite possible smallness of the combined impurity concentration and strength , it should dominate not only the @xmath7 lifetime of the gapped magnon , but also its temperature dependence in the entire low - temperature regime .
a qualitative prediction of our consideration is that @xmath72 and @xmath73 in ( [ 1gimp_est ] ) should be of the same order since both terms are related to disorder .
in addition , for samples of the same material of different quality , they must scale with the amount of structural disorder in a correlated way . in the 3d case , impurity - assisted mechanism ( [ 1gimp_est ] ) gives @xmath75 , still dominating the 3d magnon - magnon relaxation rate @xmath76 discussed above . _
experiment._the experimental part of our work is devoted to the neutron spin - echo measurements of the magnon lifetime in @xmath14 .
this material is a layered quasi-2d af with a honeycomb lattice of spin-1 ni@xmath77 ions and nel temperature @xmath78 k. a comprehensive review of the physical properties of @xmath14 is presented in @xcite .
its excitation spectrum has an optical branch with the gap @xmath79 k and an acoustic mode , as is sketched in fig .
[ fig : diagrams](d ) .
the fit of the magnon dispersion yields the following microscopic parameters : @xmath80 mev and @xmath81 mev are exchanges between first- and third - neighbor spins , and @xmath82 mev is the single - ion anisotropy .
the thermodynamic properties of bani@xmath3(po@xmath4)@xmath3 follow the 2d behavior down to @xmath83k and a small gap in the acoustic branch , @xmath84k , due to weak in - plane anisotropy is consistent with the value of the ordering temperature @xcite . for several representative spin - echo energies . ]
the spin - echo experiments were performed on the triple - axis spectrometer in22 ( ill , grenoble ) by using zeta neutron resonance spin - echo option @xcite . the incident neutron beam was polarized and the scattered beam analyzed from ( 111 ) reflection of @xmath85 heusler alloy focusing devices .
we used a fixed-@xmath86 configuration , with @xmath87 @xmath88 or @xmath89 @xmath88 .
different rf - flipper configurations were used in order to adapt the spin - echo time ( energy ) @xmath90 ( @xmath91 ) to the magnetic excitation lifetimes , typically in the range of @xmath92ps ( @xmath93ev ) . as for any spin - echo experiment @xcite , the measurement of the neutron polarization ( spin - echo amplitude ) after the scattering , @xmath94 , provides us with a direct access to the correlation function @xmath95 . for a spin - wave excitation described by a lorentzian function in energy of half width @xmath96
, one can show that @xmath97 , in which the prefactor @xmath98 depends on the spin - echo resolution . for our measurements
, we have used a @xmath99 single crystal of @xmath14 oriented with the @xmath100 and @xmath101 reciprocal axes in the scattering plane .
the spin - echo data were taken at the antiferromagnetic scattering vector @xmath102 and the energy transfer @xmath103 mev corresponding to the bottom of the dispersion curve of the gapped mode @xcite . in determining the spin - echo amplitudes ,
neutron intensities were corrected for the inelastic background , measured at the scattering vector @xmath104 and the energy transfer @xmath105 mev .
results of the temperature dependence of spin - echo amplitudes for several representative @xmath106 s are shown in fig .
[ fig : polarization ] .
solid lines are the fits of the spin - echo amplitudes with @xmath107 using relaxation rate in the functional form given by ( [ 1gmm_est1 ] ) and ( [ 1gimp_tot ] ) , @xmath108 , which we discuss next . using the full set of @xmath109 data , experimental results for @xmath110
are extracted from the fits of @xmath111 vs @xmath106 at fixed temperatures .
these results are presented in our fig .
[ fig : gamma ] together with the theoretical fits . _
comparison._the relaxation rate approaches the constant value of @xmath112 @xmath5ev at @xmath113 , in agreement with the expectation ( [ tau_0 ] ) for the gapped mode in 2d .
the low-@xmath6 dependence of the relaxation rate is following the power law much slower than @xmath57 .
the quality of the free - parameter fit of @xmath114 with just the @xmath57 law is not satisfactory for either @xmath110 or @xmath115 s in figs .
[ fig : gamma ] and [ fig : polarization ] , and the magnitude of @xmath116 also requires an unphysically large values of the magnon - magnon scattering parameter @xmath117 in ( [ 1gmm_est1 ] ) , exceeding theoretical estimates roughly tenfold . on the other hand ,
@xmath118 law gives much more satisfactory fits in the low- and intermediate-@xmath6 regime up to 12 k in both @xmath110 and @xmath115 , shown as a separate fit by the dotted line in fig .
[ fig : gamma ] .
the best fit of @xmath110 , given by solid line , is the sum of the magnon - magnon and impurity - scattering effects from ( [ 1gmm_est1 ] ) and ( [ 1gimp_tot ] ) , with the magnon - magnon and impurity - assisted parameters @xmath119 mev and @xmath120 @xmath5ev , respectively .
the same @xmath110 is used in all three curves of @xmath115 in fig .
[ fig : polarization ] , the original data from which experimental @xmath110 is extracted .
magnon bandwidth @xmath121 k and the low - energy cutoff @xmath122 k , equal to the gap in the acoustic branch , were used . of the optical magnon with @xmath123 in bani@xmath3(po@xmath4)@xmath3 .
full line is the best theoretical fit including all contributions with parameters described in the text .
dashed and dotted lines indicate separate contributions of magnon - magnon and impurity - assisted magnon - magnon scattering . ]
two remarks are in order concerning the role of the magnon - magnon relaxation rate used in fig .
[ fig : gamma ] .
first , fits of @xmath110 in fig .
[ fig : gamma ] also include a contribution from scattering off the thermally excited optical magnons , which is given by @xmath124 @xcite .
its contribution is roughly equal to that of the @xmath57-term ( [ 1gmm_est1 ] ) at @xmath125k ( @xmath126 ) , but diminishes faster at lower @xmath6 . in the fit of @xmath110
we use the value of @xmath127 @xmath5ev , about three times the theory estimate : @xmath128 @xmath5ev .
second , the theoretical estimate of the magnon - magnon interaction parameter in @xmath57 law ( [ 1gmm_est1 ] ) is @xmath129 mev , again factor 2.5 smaller than the one used in the fit ( @xmath119 mev ) . altogether , the magnon - magnon contribution to @xmath110 , shown by the dashed line and the corresponding color shading in fig .
[ fig : gamma ] , is likely a generous overestimate of its actual role in the relaxation .
still , the contribution of the impurity - assisted mechanism in @xmath110 is very strongly pronounced and is not explicable by the conventional scattering mechanisms .
for example , at 12k the impurity scattering accounts for at least 2/3 of the temperature - dependent part of @xmath110 .
the parameter of the impurity - assisted term in ( [ 1gimp_tot ] ) used in the fit is @xmath130ev , which is of the same order with the constant impurity term @xmath72 , meeting our expectations outlined above .
this is , again , the strong argument that both the constant and the @xmath6-dependent terms in the relaxation rate must have the same origin , giving further support to the consistency of our explanation of the data .
the values of @xmath73 and @xmath72 can not be determined theoretically as the impurity concentration and strength are , generally , unknown .
however , another consistency check is possible : the ratio of @xmath72 to a characteristic energy scale of the problem , @xmath131 , should give , according to ( [ tau_0 ] ) , an estimate of the cumulative measure of disorder concentration and its strength : @xmath132 .
this translates into a reasonable estimate of the disorder and its strength in bani@xmath3(po@xmath4)@xmath3 : modulation of magnetic couplings is equivalent to half of a percent of sites having @xmath22 ( @xmath21 ) of order @xmath133 ( @xmath134 ) .
the amount of structural distortion in bani@xmath3(po@xmath4)@xmath3 @xcite is consistent with the magnitude of such variations of magnetic couplings , given the strong spin - lattice coupling in this material . _
other systems._we propose that similar , and even stronger , effects of disorder in the relaxation rate must be present in the 2d noncollinear afs , in which magnon - magnon interactions acquire the so - called cubic interaction terms @xcite , absent in the collinear afs considered above .
the self - energies associated with such interaction are the same as in figs .
[ fig : diagrams](b ) and ( c ) , but with two intermediate lines instead of three . with the long - wavelength behavior of the impurity interaction to follow @xmath135 , as in the considered case , a qualitative consideration similar to ( [ 1gimp_est ] ) leads to : @xmath136 where @xmath137 , an even lower power of @xmath6 .
since the canting of spins can be induced by the external field , we propose an experimental investigation of the effect of such a field on the relaxation rate . for the 3d noncollinear afs
we predict @xmath138 .
recent neutron spin - echo experiment in a heisenberg - like af mnf@xmath3 @xcite have reported significant discrepancies between measured relaxation rates and predictions of the magnon - magnon scattering theory @xcite , precisely in the regime of low-@xmath6 and small-@xmath139 where the theory is assumed to be most reliable . although the current work concerns the dynamics of strongly gapped excitations and our results are not directly transferable to the case of mnf@xmath3
, we have , nevertheless , presented a general case in which the magnon - magnon scattering mechanism is completely overshadowed by impurity scattering , thus suggesting a similar consideration in other systems . _
conclusions._to conclude , we have presented strong evidence of the general situation in which temperature - dependence of the relaxation rate of a magnetic excitation is completely dominated by the effects induced by simple structural disorder .
our results are strongly supported by the available experimental data .
further theoretical and experimental studies are suggested .
this work was initiated at the max - planck institute for the physics of complex systems during the activities of the advanced study group program on `` unconventional magnetism in high fields , '' which we would like to thank for hospitality . the work of a. l. c. was supported by the doe under grant no .
de - fg02 - 04er46174 .
99 s. p. bayrakci , t. keller , k. habicht , and b. keimer , science * 312 * , 1926 ( 2006 ) .
t. keller , p. aynajian , k. habicht , l. boeri , s. k. bose , and b. keimer , phys .
rev .
lett . * 96 * , 225501 ( 2006 ) .
d. haug , v. hinkov , p. bourges , n. b. christensen , a. ivanov , t. keller , c. t. lin , and b. keimer , new j. phys .
* 12 * , 105006 ( 2010 ) .
b. nfrdi , t. keller , h. manaka , a. zheludev , and b. keimer , phys .
rev .
lett . * 106 * , 177202 ( 2011 ) .
a. b. harris , d. kumar , b. i. halperin , and p. c. hohenberg , phys .
rev .
b * 3 * , 961 ( 1971 ) .
s. m. rezende and r. m. white , phys .
rev .
b * 14 * , 2939 ( 1976 ) , _ ibid . _ * 18 * , 2346 ( 1978 ) . j. bass , w. p. pratt , and p. a. schroeder , rev .
mod .
phys . * 62 * , 645 ( 1990 ) ; p. l. taylor , phys . rev . * 135 * , a1333 ( 1964 ) ; s. koshino , prog .
. phys . * 30 * , 415 ( 1963 ) .
r. m. white , r. freedman , and r. b. woolsey , phys .
rev .
b * 10 * , 1039 ( 1974 ) . l. p. regnault and j. rossat - mignod , in _ magnetic properties of layered transition metal compounds _ , edited by l. j. de jongh ( kluwer academic , dordrecht , 1990 ) , p. 271 . see supplemental material at http://link.aps.org/supplemental for details of theoretical calcualtion in the honeycomb - lattice antiferromagnets .
y. kohama , a. v. sologubenko , n. r. dilley , v. s. zapf , m. jaime , j. a. mydosh , a. paduan - filho , k. a. al - hassanieh , p. sengupta , s. gangadharaiah , a. l. chernyshev , and c. d. batista , phys .
rev .
lett . * 106 * , 037203 ( 2011 ) .
g. d. mahan , _ many - particle physics _
( plenum press , new york , 1990 ) . k. hirakawa and h. ikeda , in _ magnetic properties of layered transition metal compounds
_ , edited by l. j. de jongh ( kluwer academic , dordrecht , 1990 ) , p. 231 .
e. m. lifshitz and l. p. pitaevskii , _ statistical physics ii _
( pergamon , oxford , 1980 ) , p. 133 . m.a .
kastner , r.j .
birgeneau , g. shirane , and y. endoh , rev .
mod .
phys . * 70 * , 897 ( 1998 ) ; d. c. johnston , in _ handbook of magnetic materials _ , edited by k. h. j. buschow ( elsevier science , north holland , 1997 ) .
v. g. baryakhtar and v. l. sobolev , fiz .
tverd .
tela ( leningrad ) * 15 * , 2651 ( 1973 ) [ sov .
phys .
solid state * 15 * , 1764 ( 1974 ) ] .
n. martin , l .- p .
regnault , s. klimko , j. e. lorenzo , and r. ghler , physica b * 406 * , 2333 ( 2011 ) .
f. mezei , z. physik * 255 * , 146 ( 1972 ) .
r. golub and r. ghler , phys .
lett .
a * 123 * , 43 ( 1987 ) .
n. martin , l .- p .
regnault , and s. klimko , j. phys .
: conf . ser . *
340 * , 012012 ( 2012 ) .
a. l. chernyshev and m. e. zhitomirsky , phys .
rev .
lett . * 97 * , 207202 ( 2006 ) .
* lifetime of gapped excitations in collinear quantum antiferromagnet : + supplemental information * + 0.35 cm a. l. chernyshev@xmath140 , m. e. zhitomirsky@xmath141 , n. martin@xmath141 , and l .- p .
regnault@xmath141 + 0.15 cm _
@xmath142department of physics , university of california , irvine , california 92697 , usa + @xmath141service de physique statistique , magntisme et supraconductivit , + umr - e9001 cea - inac / ujf , 17 rue des martyrs , 38054 grenoble cedex 9 , france _
+ ( dated : june 20 , 2012 ) + -0.1 cm |
the ` bright nova ' s andromedae ( s and ) was discovered in the central bulge of m31 ( ngc 224 ) during late august of 1885 .
it is believed to have been a type ia supernova ( sn ia ) on the basis of an apparent absence of hydrogen balmer lines @xcite and a close match of nearly all reported spectral features to those of sne ia @xcite .
however relative to normal sn ia events , sn 1885 was subluminous with an unusually rapid initial decline and a somewhat redder @xmath14 color @xcite . a recent analysis of sn 1885 observations by @xcite suggests a decline rate of @xmath15 mag and a maximum brightness of @xmath16 and @xmath17 , assuming a m31 distance modulus of ( @xmath18)@xmath19 = 24.4 corresponding to a distance of 760 kpc .
the remnant of sn 1885 remained undetected for over a century despite numerous searches in the optical and radio @xcite .
the remnant was finally detected in 1989 , not through its emission , but as a small dark spot of absorption in the m31 bulge on an image taken using a near - uv filter ( 3900 @xmath20 100 ) sensitive to h & k resonance line absorption @xcite .
although @xcite report a possible weak radio detection at 8.4 ghz , sn 1885 continues to be essentially unobservable via emission , with recent deep _ chandra _ x - ray imaging of the m31 bulge showing no associated x - ray source at its position @xcite .
following its ground - based detection , optical and uv imaging and spectra of the sn 1885 remnant ( hereafter referred to as `` snr 1885 '' ) with instruments aboard the _ hubble space telescope _
( _ hst _ ) revealed a @xmath21 @xmath22 diameter absorption patch produced principally by ca expanding at velocities up to 13,000 km s@xmath7 @xcite .
the observed low flux at the h & k blended line center indicated a foreground starlight fraction of around 0.20 , placing snr 1885 some 65 pc to the near side of the m31 bulge midpoint , comparable to its projected 55 pc (= @xmath23 ) distance from the m31 nucleus .
other absorption lines detected in snr 1885 included @xmath244227 , @xmath243441 , and @xmath243720 . a weak imaging detection of the remnant with the _ hst _ wide - field planetary camera-2 ( wfpc2 ) in the wavelength range 2200 to 3000 , which is rich in strong lines , suggested a depth of absorption consistent with being fully saturated and an iron mass of @xmath25 @xmath26 @xcite . the fortuitous positioning of a relatively young and nearly freely expanding sn ia remnant silhouetted against m31 s bright central bulge offers an unique opportunity to map the distributions of ca and fe rich debris in a sn ia .
in this paper , we present images of snr 1885 taken with the advanced camera for surveys ( acs ) on _ hst _ at a significantly higher angular resolution and signal - to - noise - ratio than previous images .
the acs images show a sufficiently well - resolved remnant to distinguish between various proposed sn ia explosion scenarios .
the observations , image results , our modeling results , future observations , and conclusions are given in @xmath27 , respectively .
high - resolution optical and ultraviolet images of snr 1885 , detected via resonance line absorptions of m31 bulge starlight , were obtained between 2004 august and november using the advanced camera for surveys ( acs ; @xcite ) aboard _
acs has two main imaging configurations and both were used .
the acs wide field channel ( wfc ) consists of two @xmath28 ccds providing a field of view @xmath29 with an average pixel size of @xmath8 , while the acs high resolution channel ( hrc ) consists of a single @xmath30 ccd providing a spatial resolution of @xmath31 per pixel and a nominal @xmath32 field of view .
the acs / wfc was used in combination with narrow passband ` ramp ' filters to image snr 1885 in three separate wavelength ranges .
these were the resonance absorption line of ca i 4226.73 , the ca ii h & k doublet at 3933.66 and 3968.47 , and a continuum band free of strong remnant absorption lines centered at 4600 .
default acs / wfc and acs / hrc dither ` box ' patterns were used which are 4-point parallelogram patterns with relative pixel coordinates ( 0 , 0 ) , ( 5.0 , 1.5 ) , ( 2.5 , 4.5 ) , ( -2.5 , 3.0 ) , thus having both integer and sub - pixel shifts . a image of snr 1885 was taken with the 2% ramp filter fr423n centered at 4227 , resulting in an effective bandpass of 75 fwhm and a peak filter transmission of about 60% .
eight 1120 s exposures were taken during four _ hst _ orbits on 2004 november 2 in a four - point dither pattern for a total exposure time of 8960 s. a h & k absorption image of snr 1885 was obtained with the acs / wfc 2% ramp filter fr388n centered at a wavelength 3950 , chosen to match the center of the remnant s broad ca ii absorption line profile as observed in the _ hst _ faint object spectrograph ( fos ) spectrum @xcite . at this wavelength ,
the filter s bandpass is 100 fwhm with a peak transmission of about 50% .
images were taken on 2004 november 8 with a dithering pattern and exposure times identical to those of the ca i images .
a continuum acs / wfc image was also taken to allow for correcting small - scale variations in the distribution of background and foreground starlight from the m31 bulge at the position of snr 1885 .
the continuum image was taken with the 9% ramp filter fr459 m centered at a wavelength 4600 , giving a bandpass of about 350 fwhm and a peak transmission of nearly 80% .
the spectral region is adjacent to the and lines but free of any strong line absorption . due to to the filter s
wide bandpass and the increasing luminosity of m31 bulge stars to longer wavelengths , a relatively short exposure was adequate and dithered images were taken during just two orbits on 2004 august 25 for a total exposure time of 4666 s. for obtaining uv and images , we used the available broadband filters on the higher resolution acs / hrc . for the image , we selected the f330w filter , which peaks at 3300 and extends over 29753700 .
the filter catches the resonances lines at 3021 , 3441 , and 3720 @xcite , but most of its coverage is line - free continuum .
consequently , the filter provided only a weak detection of .
indeed , we used the image in part as a background continuum image to help correct the and images ( [ cai ] , [ caii ] ) . dithered f330w images were taken over three orbits on 2004 october 13 for a total exposure time of 7800 s. for the image , we selected the f250w filter , which peaks in transmission at 2500 but has a broad transmission passband extending over @xmath33 .
the filter encompasses the resonance lines at 2343 , 2373 , 2382 , 2586 and 2599 , but it also includes lines at 2524 and 3021 , the line at 2852 , and the doublet at 2796 and 2803 , all of which are predicted to have significant optical depths @xcite . dithered f250w images were taken over 6 orbits on 2004 november 20 for a total exposure of 15,600 s. standard acs pipeline iraf / stsdas data reduction was done , including debiasing , flat - fielding , geometric distortion corrections , photometric calibrations , and cosmic ray and hot pixel removal .
the stsdas _ drizzle _ task was used to combine exposures in each filter .
figure [ fig : ca2_vs_offband ] shows a @xmath34 log scaled intensity view of snr 1885 and its local environs within the bulge of m31 .
the h & k image ( _ upper panel _ ) reveals the remnant of sn 1885 as a dark and circular spot at the historically reported location of sn 1885 .
the coordinates for the center of the remnant are : @xmath35 , @xmath36 .
in contrast to the image , the offband 4600 fr459 m continuum image ( fig . 1 , _ lower panel _ ) shows no appreciable absorption at the position of snr 1885 .
because sn 1885 s projected location is uncontaminated by the dust lanes that thread the m31 bulge , the lack of appreciable absorption seen in the 4600 image at the position of snr 1885 suggests little internal dust formed in this sn ia s metal - rich ejecta .
figure [ fig : all_four_species ] shows all four of the individual absorption images of snr 1885 : @xmath244227 ( _ upper left _ ) , h & k @xmath373934,3968 ( _ upper right _ ) , @xmath373441,3720 ( _ lower left _ ) , and @xmath372358,2599 plus @xmath242800 ( _ lower right _ ) .
the images shown here are in their native spatial pixel scales ( acs / wfc : @xmath8 ; acs / hrc : @xmath31 ) and are raw in the sense that they have not been corrected for any background variations in the surface brightness of the m31 bulge .
the surface brightness of starlight from the bulge of m31 has both a large scale gradient and smaller scale fluctuations . to correct the and absorption images of snr 1885 for these non - uniformities , we used combinations of the offband 4600 continuum image and the ` ' f330w filter image , in the ratio 4:1 and 3:1 , respectively for the and images .
as noted above in @xmath382 , the f330w filter is broad and , although it encompasses a number of resonance lines , it mostly covers unabsorbed continuum .
we judged it better to use a background constructed from continuum both longward ( 4600 ) and shortward ( f330w ) of the and lines , and to accept the mild contamination from absorption in the f330w continuum , than to use the 4600 image alone . because the acs / hrc f330w image has a native image scale of @xmath31 per pixel
, we first resampled the acs / wfc , , and 4600 images to @xmath39 per pixel using the iraf image interpolation task _ magnify _ , before applying the background correction .
figures [ fig : cai_orig_vs_new ] and [ fig : orig_vs_new ] show in close - up the resampled , background - corrected images of snr 1885 in and absorption , along with the original , uncorrected images .
we discuss these images in more detail below in [ cai ] and [ caii ] .
the distance to m31 is estimated to be between 760 and 790 kpc based on studies of m31 cepheid and rr lyrae variables , globular clusters , and stars at the tip of the red giant branch ( @xcite and references therein ) . in the discussions below ,
we have adopted a distance of @xmath3 kpc to m31 and the snr 1885 .
[ [ cai ] ] figure [ fig : cai_orig_vs_new ] shows the and 4600 continuum images of snr 1885 resampled to @xmath39 per pixel .
it also shows two slightly different background - corrected versions of the image . like that seen in the raw image presented in figure 2 , the uncorrected but resampled ca i absorption image has a markedly lopsided ` c ' shaped appearance with the strongest absorption oriented approximately in the direction of the nucleus of m31 toward the northeast .
the remnant s overall diameter in this image is roughly @xmath40 corresponding to a linear diameter of 2.5 pc at a distance of 785 kpc .
the two background - corrected images shown in figure [ fig : cai_orig_vs_new ] are still noticeably lopsided , though much less so , and both appear more complete in the sense of showing a greater degree of absorption near the center and along the western limb . in the first background - corrected version ( fig . 3 , _ lower left _ ) , we used only the 4600 continuum image to remove bulge starlight variations , while for the second version ( fig . 3 , _ lower right _ ) which we regard as being better
, we used a combination of 4600 and f330w images for the background in the ratio 4:1 , respectively .
one can see that the bulge non - uniformities are significantly decreased in both corrected images .
moreover , the good agreement between the two background - corrected ca i images indicates that the mild contamination from absorption in the f330w image does not have a substantial effect .
while our use of adjacent continuum images should largely remove small scale structures attributable to variations in the background starlight , we can not entirely remove m31 bulge light variations because some 25% of the bulge starlight lies foreground to snr 1885 ( see [ caii ] below ) .
nonetheless , the deepest levels of absorptions appear along the remnant s eastern and northern limbs .
[ [ caii ] ] in the _ hst _ acs images , snr 1885 is most prominently visible through h & k line absorption .
figure [ fig : orig_vs_new ] shows the acs / wfc image both in its original @xmath8 pixel scale format and then after resampling to @xmath39 per pixel plus background corrected using the 4600 and f330w images combined in the ratio 3:1 .
the intensity toward the center of the uncorrected image falls to about 30% of the background intensity of starlight , making snr 1885 the darkest feature in the inner m31 bulge at 3950 .
both original and background corrected images indicate the likely presence for a few absorption ` clumps ' having an angular size around that of the acs / wfc image resolution ( @xmath8 ) , @xmath41 pc at a distance of 785 kpc .
this scale suggests a span in velocity for such clumps of around @xmath42 km s@xmath7 .
the most prominent absorption clump is about 25% darker than its surroundings and lies some @xmath9 due north of the remnant s absorption center .
while some of the small - scale absorption structures seen in the image are likely due to variations in both the corrected bulge starlight background and in the starlight lying in front of the remnant , the strength and non - randomness of the darkest features suggests the existence of a broken ring of -rich clumps at a radius of @xmath5 ( @xmath43 ) located along the inner edge of a limb - brighten shell . aside from these apparent clumps ,
the degree of absorption in the image appears approximately constant from the center of the remnant out to a radius of @xmath5 , corresponding to a free expansion velocity of @xmath44 .
this suggests that the absorption is at least partially saturated in the central region , as previously concluded from the fos spectrum of snr 1885 discussed by @xcite .
the raw and background - corrected images of snr 1885 also indicate a fairly spherical remnant structure .
the spherical shape and outer extent of the remnant s absorption is illustrated by the @xmath45 diameter circle shown in figure 4 .
this circle encompasses virtually all of the observed absorption .
the maximum detected dimensions of the sn 1885 remnant as seen in both the and images are shown in figure [ fig : ca1_vs_ca2 ] .
the image shows that absorption extends out to a diameter of @xmath46 , which is larger than that seen via absorption ( diameter = @xmath47 ) .
comparison of these and images which are shown here greatly stretched to bring out absorption in the outermost parts of the remnant , shows that the image as a whole appears offset relative to the center of the image , by approximately @xmath8 in the direction of the nucleus of m31 .
the offset is consistent with the asymmetric appearance of the image , and probably reflects a real intrinsic asymmetry in the distribution of versus .
the stretched image shown in figure 5 indicates an outer radius for absorption of @xmath48 at m31 s estimated distance of @xmath49 .
this is equivalent to an expansion velocity of @xmath50 at the @xmath51 age of the remnant .
this velocity estimate is in agreement with the maximum velocity of @xmath52 directly measured from a fos spectrum obtained by @xcite .
this agreement supports the notion that the absorbing -rich ejecta are in free expansion .
moreover , since the image probes the size in the transverse direction while the spectrum probes the size in the line - of - sight direction , the similarity of the two expansion velocities indicates that the ejecta structure is approximately spherical . finally , as noted above the average central intensity in the image
is about @xmath53 that of the local bulge intensity of starlight .
this is somewhat higher than the average central intensity of @xmath54 seen in the fos spectrum of snr 1885 @xcite .
the slightly brighter average central intensity measured from the acs / wfc image compared to the fos spectrum suggests that the foreground starlight fraction may be about @xmath54 , higher than the @xmath55 fraction found by @xcite from a best - fit model of the fos spectrum .
a possible reason for this slight discrepancy in central intensities is that center of the @xmath56 circular aperture of the fos just happened to fall on one of the darker patches of absorption ( see fig . 4 )
. [ [ fei ] ] the remnant s absorption distribution was imaged using the broad , near - uv acs / hrc filter f330w . while this filter s bandpass is sensitive to three lines ( 3021 , 3441 , and 3720 ) , the two strongest lines , at 3021 and 3720 ,
are detected only partially and at low filter transmission levels ; that is , only the redshifted absorption of the 3021 line and just the blueshifted side of the 3720 line .
in addition , the f330w filter covered a considerable continuum region between @xmath57 outside of the lines and , in fact , was used to help compose background bulge starlight images for the and acs / wfc images .
nonetheless , faint absorption was detected on the f330w image .
the image ( fig .
2 ) resembles most the uncorrected image , showing a lopsided and broken absorption patch some @xmath58 in diameter , a break along the southwest limb , and the strongest absorption concentrated toward the northeast .
the image also shows a peak of absorption roughly @xmath59 pixels across ( @xmath60 ) , i.e. , about the same size as that for the clumps , displaced toward the northeast by about @xmath9 from the center of the absorption patch .
[ [ feii ] ] m31 bulge starlight is relatively faint between @xmath61 and faintest between 2300 2600 where several strong resonance lines are found .
this fact , together with the rising continuum flux of bulge starlight toward the redward portion of the @xmath62 passband of the f250w filter on acs / hrs resulted in a low signal - to - noise 3000 continuum dominated image .
this , in turn , produced a very weak detection of snr 1885 presumably due to line absorption ( fig . 2 , _ lower right panel _ ) based upon ionization models which predict that the remnant should be almost black over the @xmath63 range @xcite .
notwithstanding its comparatively poor detection of snr 1885 , the f250w image bears a clear resemblance to the image of the remnant .
this can be seen in figure 6 were we show a side - by - side comparison of snr 1885 in and the f250w images .
it is important to note that the f250w filter covers not only lines but also the mg ii doublet at @xmath64 and several lines ( see @xmath383.3 ) .
however , the similarity of the f250w image to the ca ii image but not to the or images ( see fig . 2 ) suggests that does not make a major contribution to the f250w image . while @xmath64 could also contribute to the detection of snr 1885 on the f250w image , mg ii
is expected to lie outside of the ca - rich ejecta region and should thus lead to a larger diameter appearing remnant .
since this is not what is seen in the f250w image , we conclude that the f250w image largely shows the distribution of -rich ejecta in snr 1885 .
in this section , we compare our acs imaging data of snr 1885 to off - center delayed - detonation models of type ia sne ( sne ia ) .
there is general agreement that sne ia result from processes involving the combustion of a degenerate c - o white dwarf ( wd ) that somehow reaches the chandrasekhar mass limit , @xmath65 . at present ,
two mechanisms are considered most likely to bring a wd to this point . currently , the leading idea is that a single wd accretes matter by roche - lobe overflow from a main sequence or red giant companion @xcite .
an alternative possibility is the merger of two low - mass wds in a binary @xcite .
either way , the explosion is triggered near the center by compression .
one of the main uncertainties in sn ia physics is how the nuclear burning flame propagates through the white dwarf .
specifically , does it propagate as a weak supersonic detonation front , or as a subsonic deflagration wave ?
whereas detonation fronts are fairly well understood , modeling the propagation of a deflagration wave is difficult because the effective speed of deflagration burning depends on details of the development of multi - dimensional instabilities which , in turn , depend sensitively on the initial conditions that lead to the thermonuclear runaway .
despite significant recent progress , direct deflagration modeling represents a major computational challenge .
this is because of the multi - dimensional physics involved , the wide range of scales spanning 5 to 6 orders of magnitude , and the sensitivity to initial conditions . as a result ,
fully self - consistent computer models are beyond reach at the present time , and additional observational constraints are needed to make progress on the subject .
for example , observational knowledge of the distribution of the products of nuclear burning , as in snr 1885 , can constrain the conditions under which burning must have taken place .
pure detonation explosions of @xmath65-mass models have long been ruled out for normal or subluminous sne ia .
such models predict that the wd is incinerated almost entirely to @xmath66 , contrary to observed sn ia spectra which show a range of intermediate mass elements .
pure deflagration explosions also appear to be ruled out for normal sne ia .
three - dimensional numerical computations show that deflagrations produce deep rayleigh - taylor unstable plumes , which lead to extensive radial , macroscopic mixing of burning products @xcite . as a consequence ,
pure deflagration models predict a chemically mixed region of burning products , surrounded by a thick layer of unburned c - o of several tenths of a solar mass .
by contrast , spectra of normal sne ia reveal a chemically stratified structure .
moreover , there is no evidence for rising deflagration plumes whose signatures should be visible in infrared spectra of normal sn ia at late times or in subluminous sne ia such as sn 1999by at early times @xcite .
better agreement with observations is obtained with so - called ` delayed - detonation ' ( dd ) models in which the explosion is assumed to begin as a deflagration which then ( somehow ) develops into a detonation .
the mechanism by which a deflagration might turn into a detonation is not well understood .
thus simulations typically make the _ ad hoc _ assumption that at some point a deflagration wave simply turns into a detonation , and the density at which this transition takes place is treated as a free parameter .
increasing the transition density causes the transition from deflagration to detonation to occur earlier , increasing the amount of @xmath66 synthesized .
variation in this one parameter can account for the observed gross variation in the properties of sne ia : how much @xmath66 is produced , the overall distribution of elements , and the brightness of the sn ia event @xcite .
arguments about sne ia in general do not necessarily apply to an individual object such as sn 1885 .
however , delayed - detonation models can reproduce both early to late time light curves and the spectra of both normal and subluminous sne ia light spectra in the optical to the ir @xcite .
we therefore base our comparison to observations of snr 1885 on a suite of spherical delayed - detonation models @xcite .
the models start from the progenitor , 5p01z22 , a c - o white dwarf taken from the core of an evolved 5 m@xmath67 main sequence star with solar metallicity . through accretion
, this core approaches the chandrasekhar limit .
a deflagration begins spontaneously when the core reaches a central density of @xmath68 and the wd mass is close to @xmath69 @xcite .
as mentioned above concerning dd models , the density at which the deflagration is assumed to transition into a detonation is the key parameter that produces variation in sne ia . for extragalactic sne ia , the transition densitycan
be adjusted so as to match the observed light curve , in particular its rate of decline @xcite .
the chemical profile of the model then predicts a spectrum that can be compared to observation . in the case of snr 1885
, we will instead choose the transition density so as to reproduce the chemical profiles observed with acs .
the maximum brightness and light curve then become a prediction of the model .
the _ hst _ acs images of snr 1885 presented in [ imageresults ] show at velocities up to @xmath70 . by comparison
, dd models can produce ca - rich ejecta with maximum velocities that range from @xmath71 for a very subluminous sn ia resembling sn 1991bg , up to @xmath72 for normal sn ia . the specific model that best matches snr 1885 is 5p01z22.16 , in which the ` .16 ' refers to a dd transition density of @xmath73 for the progenitor 5p01z22 .
in this model , the deflagrated region encloses @xmath74 , and the explosion produces a net total of @xmath75 of @xmath66 . model 5p01z22.16 , which also provides a good fit to the subluminous sn ia object sn 1986 g @xcite , is subluminous by about @xmath76 magnitudes , predicts a peak brightness of @xmath77 and @xmath16 , and a 15-day decline rate of @xmath78 and @xmath79 .
this peak brightness in @xmath80 agrees well with that observed for sn 1885 , but the decline rate is slower than the @xmath81 reported for sn 1885 @xcite .
a possible explanation of the discrepancy in the decline rate may be related to uncertainties in the time @xmath82 of maximum light of sn 1885 .
all sne ia show a flat maximum , followed by a phase of increasing rate of decline @xcite .
a shift of @xmath83 days in @xmath82 would bring the model into closer agreement .
in constructing off - center dd models , we have followed the prescription of @xcite . in this prescription ,
the initial deflagration phase is modeled assuming spherical symmetry .
the deflagration begins at the center , and propagates outward in a subsonic deflagration wave .
the energy deposited by deflagration causes the entire white dwarf to expand .
when the density at the leading edge of the deflagration wave has fallen to a certain transition density , a detonation is ignited by hand at a single point ( the north pole ) of the deflagration front .
the simulation is then continued in 2 dimensions assuming cylindrical symmetry .
we choose to impose spherical symmetry on the initial deflagration phase for two reasons .
the first reason is simply computational tractability : 3-dimensional simulations are computationally expensive , requiring many @xmath84 cpu hours @xcite , and indeed no fully consistent dd models have ever been calculated .
the second reason is that multi - dimensional simulations tend to predict mixed chemical profiles , which are at odds with observations both of typical sn ia ( see [ scenarios ] above ) and of snr 1885 , where the observed image indicates some radial stratification .
no mixing seemed to us to offer a more promising approach than full mixing .
however , in [ modelimages ] below we will show that the ca image of snr 1885 is more filled - in at its center than predicted by the model , suggesting that the assumption of spherical symmetry over - corrects the problem , producing more stratification than needed .
we remedy this _ a posteriori _ in [ modelimages ] , by taking the final , fully - exploded wd , and partially remixing its central regions .
the 3d calculations of @xcite show that detonation proceeds mainly along radial fingers , leaving behind connected regions of unburned matter . to allow for this
, we assumed that the deflagration burns only about @xmath85 of the matter , leaving the remaining matter unburned .
the unburned material plays an important role in that it allows the subsequent detonation to move back into and through the center of the partially deflagrated core , which offers a better approximation to the 3d calculations of @xcite than models in which there is no unburned material , and the detonation front is forced to circulate around the core @xcite .
deflagration burning causes the entire wd to expand .
several seconds into the deflagration , when the density has decreased by about two orders of magnitude , we initiate the detonation at a point on the boundary of the deflagration wave .
we treat the density of the transition as a free parameter , and we characterize its location by the mass @xmath86 interior to the detonation point . at the same time that we initiate the detonation , we switch from spherical to cylindrical symmetry , following the explosion on a grid of @xmath87 cells in the radial and angular directions .
the detonation begins at the pole of the cylindrically symmetric configuration , and propagates away from the pole in all directions .
the detonation front moves not only outward and sideways through the unburned wd , but also inward through deflagrated core , consuming the @xmath88 fraction of unburned material there . in the model , the detonation front reaches the surface of the wd at the pole about @xmath89 seconds after the detonation , and it reaches the surface at the opposite pole about @xmath90 seconds after that , consistent with the times obtained in the 3d calculations of @xcite ( who did not , however , continue their calculation to homologous expansion ) . over the next 5 to 10 seconds , the hot wd accelerates into free expansion , and the stored thermal energy is converted into kinetic energy .
the density structure of the expanding ejecta freezes out as the sound speed plummets .
about one second after entering the 2d phase , we switch back to 1d hydrodynamics , evolving each of the @xmath91 angular segments as if it were spherically symmetric . at this time , angular gradients in the ca - rich layers are less than 2% .
all numerical calculations were done using the hydra code ( see @xcite ) .
hydra consists of a number of modules that solve simultaneously the hydrodynamical equations , nuclear and atomic networks , and radiation transport equations . for spherical geometry ,
the hydro equations are solved in the lagrangian frame , and include a front tracking scheme to resolve shock fronts @xcite . for cylindrical geometry ,
the coupled hydro and gravitational ( poisson ) equations are solved in an eulerian frame using a code based on prometheus @xcite without adaptive mesh refinement .
the hydro modules use the explicit piecewise parabolic method ( ppm ) of @xcite to solve the compressible reactive flow equations .
hydra does not include a detailed model of the nuclear flame and of the associated development of instabilities .
instead , the code parameterizes the deflagration speed semi - analytically , the parameters being adjusted so as to reproduce deflagration speeds obtained in 3d calculations . @xcite
have shown that , in models with a parameterized deflagration speed , the abundance of chemical elements synthesized is relatively insensitive to the deflagration speed , but rather depends mainly on the amount of electron capture that takes place during deflagration , which in turn depends on the total amount of material burned during the deflagration phase .
for the nuclear equation of state and the reactions , hydra uses the nuclear reaction - network library of ( * ? ? ? * and references therein ) .
we take into account 218 isotopes for all models . for spherical geometry
, we use 912 radial layers , while for cylindrical geometry we adopt a relatively modest resolution of @xmath87 cells .
computationally , there is a trade - off between a large nuclear network and high spatial resolution .
we opt for a large nuclear network ( rather than the 4 or 5 isotopes typically considered in 3d calculations ) because numerical tests indicate that the final chemical composition is more sensitive to a reduction in the nuclear network than to a factor 2 reduction in the spatial resolution @xcite . in any case
, the detonation phase is relatively brief and occurs on an already expanding wd so the effects of moderate spatial resolution during detonation are not expected to be dramatic .
still , there is certainly room for improvement in this aspect of the modeling .
figure [ chem ] shows the late - time chemical distribution in model 5p01z22.16 , with transition mass @xmath92 .
this figure shows the distribution in velocity of o ,
si , ca , fe , and @xmath66 several minutes after the explosion , which is well into the phase of free , homologous expansion .
the detonation axis points upward .
since the @xmath66 in snr 1885 today has long since decayed radioactively to @xmath93 , the figure also shows the distribution of combined @xmath94 . the chemical structure seen is similar for models with transition mass @xmath92 to @xmath95 . during the early stage of the deflagration ,
densities are high enough , @xmath96 , that electron capture can take place .
this produces a central region containing stable isotopes of fe , co , and ni with a mass fraction for fe of @xmath97 .
as the density drops in the expanding wd , electron capture ceases and the later stages of deflagration yield about @xmath98 of radioactive ni and ca .
the detonation phase produces off - center layers of explosively synthesized isotopes , from @xmath66 and ca on the inside to oxygen on the outside .
the element abundances are characteristic of those obtained in various stages of burning .
the outermost oxygen is partially burned , its mass fraction having increased from the 50% of the original c - o wd to about 70% .
the distribution of ca in model 5p01z22.16 is qualitatively similar to that seen in the hst / acs images of snr 1885 .
the model shows ca in a lopsided shell with a mass fraction around 3% from about 3500 to @xmath99 , peaking at a mass fraction of about 6% at @xmath100 .
the velocity of the peak increases slightly to @xmath101 if the transition mass is increased to @xmath102 .
the final fe distribution has an inner region of about 50% , produced during the early deflagration .
this region is surrounded by a lopsided shell , produced by detonation to @xmath66 , where the mass fraction rises to almost 90% .
the mass fraction decreases slowly outward , declining to about 1% at @xmath103 .
the _ hst _ imaging observations of snr 1885 reveal not the density itself , but rather the column density , i.e. , the density projected along the line - of - sight . thus figure [ rho ] shows the column densities of ca and fe for model 5p01z22.16 , with the detonation axis taken at various inclinations to the line - of - sight .
the projected distributions appear round when seen face - on ( @xmath104 ) but they become asymmetric at higher inclinations . the asymmetry remains modest even at high inclinations , in part because the intrinsic asymmetry is not large and in part because projection tends to smear out the asymmetry .
comparison between model and observed images of snr 1885 does not tightly constrain the inclination .
however , the observed image is clearly asymmetric , and we adopt an inclination angle of @xmath105 because this angle is consistent with the projected asymmetry observed .
@xcite and @xcite argued that the main source of ionization in snr 1885 is photoionizing uv from the bulge of m31 , and that the main source of uv opacity in sn 1885 is from resonance lines of neutral and singly - ionized species .
recombination and charge exchange are negligible in snr 1885 120 years after the explosion .
@xcite also pointed out that if the ejecta were optically thin , then the lifetimes of and exposed to uv light from the bulge would be quite short ; @xmath106 and @xmath107 , respectively .
the errors quoted come from uncertainty in the line - of - sight location of sn 1885 within the bulge , but there could be a further factor of two uncertainty from photoionization cross - sections and from the extinction correction to the photoionizing uv flux .
predicted photoionization lifetimes as short as these pose a problem .
although resonance lines can block about half the photoionizing uv light at the present time , that still leaves half the uv to get through and reduce the and below the currently observed levels .
@xcite suggested that and might be present , in spite of their short optically thin photoionizing times , if the remnant were optically thick in neutral continua , notably , earlier in its history .
they did not , however , carry out the detailed model computations necessary to demonstrate this possibility . in the present paper ,
we have carried out the needed time - dependent , photoionization computations , and these show that the observed level of ionization in snr 1885 is entirely consistent with model expectations .
we computed the photoionization of snr 1885 as follows .
we assumed an incident uv spectrum following the _ fos _ data of @xcite down to 2221 , from the _ iue _ data of @xcite down to 1225 , and from the _ hut _ data of @xcite down to 912 , dereddened using the @xcite extinction curve with color excess @xmath108 @xcite and @xmath109 .
we treated the photoionizing uv spectrum incident on snr 1885 as being isotropic .
@xcite had noted that snr 1885 should be receiving more radiation on the side facing the nucleus of m31 and suggested that for this reason snr 1885 should be more ionized on that side .
however , the acs images show precisely the opposite . not only do the images show concentrated toward the bulge , but an _ fos _ spectrum of the remnant showed redshifted by @xmath110 , corresponding to the side closer to the bulge @xcite .
thus an anisotropic photoionization model can not account for the observed asymmetry in the distribution .
rather , the observed asymmetry must reflect an intrinsic asymmetry in the explosion , such as that produced by an off - center delayed - detonation . for simplicity , therefore , we neglected any anisotropy of the incident photoionization .
the evolution of the ionization structure of c , o , mg , si , s , ca , mn , cr , fe , and ni was computed starting from about 10 years after the explosion .
we took the initial ionization state to be mainly neutral , with an ionization fraction of 1% in the central deflagrated region where there is no radioactive heating , and 10% in the detonated region where there is radioactive heating , dominated initially by @xmath66 @xmath111 @xmath112 @xmath111 @xmath93 and later by @xmath113 @xmath111 @xmath114 @xmath111 @xmath115 .
although this initial degree of ionization is plausible , it is not well - determined _ a priori _ by the physics because of complications associated with non - local energy transport by fast electrons in the presence of magnetic fields @xcite .
tests indicated that the degree of ionization at the present time is most sensitive to the initial ionization fraction in the @xmath66 detonation layer .
reducing the ionization in this layer to 1% can delay ionization of and by about 30 years .
the initial ionization state adopted here was chosen because it is reasonable and because it reproduces the observed degree of ionization at the present time .
we computed the temperature structure by taking into account adiabatic expansion , radioactive heating , and the uv radiation field .
the electron temperature affects recombination rates and hence the ionization balance at early times when the density is high .
line opacities were computed in the sobolev approximation for homologously expanding atmospheres , omitting stimulated emission . in homologously expanding ejecta , the radius equals velocity times age , @xmath116 , and the column density per unit velocity @xmath117 of any species is equal to its density @xmath118 times the age @xmath119 . in the nomenclature of @xcite , the sobolev optical depth @xmath120 at wavelength @xmath24 for a bound transition with absorption oscillator strength @xmath121
is @xmath122 where @xmath123 and @xmath124 are the electron charge and mass , @xmath125 the speed of light , and @xmath119 is the 120yr age of the remnant .
besides resonance lines , we also took into account excited lines with lower levels originating from the ground level .
we computed the level populations in a 2-level approximation in which the lower levels are excited by ( optically thin ) ir and optical radiation from the bulge and which then decay by spontaneous emission .
altogether , we included 6,748 bound - bound transitions with wavelengths , energy levels , and oscillator strengths taken from @xcite . both to follow photoionization and to take into account shielding by neutral atoms , we included bound - free opacities from the ground state for all neutral isotopes of the elements above .
photoionization cross sections were taken from @xcite to which we added some resonances treated in the narrow line approximation based on data from the opacity project @xcite .
finally , we also took into account the doppler shift of the continuum .
the final result of all these calculations is that shielding of uv radiation by bound - free transitions of , , and proved to be the key in delaying photoionization so that and can survive in snr 1885 up to the present time .
all the models showed that ca and fe is mostly neutral at early times , but that and will rapidly disappear as soon as photoionizing uv radiation can penetrate the ejecta . without bound - free shielding ,
is depleted to 10% after about 50 years , whereas with shielding , photoionizing radiation remains blocked for almost a century .
whereas the uv opacity from bound - free transitions evolves substantially , most of the bound - bound transitions remain optically thick over the life of snr 1885 up to the present time , so that the uv line - blocking remains almost constant . figure [ ion ] shows the evolution of the /ca fraction .
the high - velocity , outermost layers having a solar - like abundance of ca are quickly photoionized , whereas the inner , ca - rich layers remain opaque for about a century .
as previously argued by @xcite , ca - rich ejecta are currently undergoing a phase of rapid photoionization to .
indeed , our models predict that absorption will become unobservably small in just another 30 years or so , when the remnant is roughly 150 years old .
figure [ image ] shows model , , and images that can be compared directly to the acs images .
these images take into account not only the chemical and ionization structure of the remnant , as described above , but also the transmission curves of the _ hst _ filters , the contribution of different lines to absorption , and the finite angular resolution of _ hst_. as discussed in [ chemical ] , the detonation axis in the model is taken to be inclined at @xmath105 to the line - of - sight . the model predicts the image to be insensitive to the precise ionization structure , since ca is mostly singly - ionized .
the model also predicts that the ca image should have a sharp edge , where the mass fraction of ca drops rapidly from 3% to the pre - explosion , solar - like abundance .
the images shown in figure [ image ] are for model 5p01z22.16 , which as described in [ ddchoice ] was chosen so as to reproduce the maximum velocity of observed in snr 1885 .
the model image in figure [ image ] is as seen through the _ hst _ ramp - filter fr388n centered at @xmath126 and @xmath127 wide , and takes into account the doublet character of the h & k absorption .
the model predicts that the optical depth should exceed unity in a ring from @xmath128 to @xmath129 , which are close to the values measured for snr 1885
. the model and images in figure [ image ] appear off - center by about @xmath10 and @xmath130 , respectively , again in good qualitative agreement with the observed images .
the offset is caused primarily by the asymmetric chemical distribution produced by the assumed off - center delayed - detonation .
the asymmetry is enhanced by the self - shielding of denser regions against photoionization .
a defect of the model is that the predicted and images are emptier at their centers than observed in snr 1885 .
it is natural to attribute this defect to our assumption that the early deflagration phase of the explosion was spherically symmetric , which artificially suppresses radial mixing by rayleigh - taylor instabilities in the deflagration wave .
indeed , the observed images show structure down to scales @xmath131 comparable to the angular resolution of the observations , such as might perhaps be produced by instabilities . to remedy this defect of the model , we tried radially re - mixing the final chemical profile in various ways
. figure [ mix ] shows the effect on the model and images of radially remixing of : ( a ) layers within the central deflagrated region ; ( b ) layers within the outer boundary of the ca layer ; or ( c ) the entire remnant .
all three cases fill in the central ca hole in the model to a greater or lesser degree .
the observed ca images of snr 1885 are intermediate between cases ( a ) and ( b ) , suggesting that there may have been a moderate amount of radial mixing between the inner deflagrated region and the surrounding detonated layers .
this level of mixing is less than that predicted by current 3d calculations of rt - dominated deflagration fronts @xcite .
smaller amounts of mixing could result if the initial conditions at the onset of runaway burning produce a less violent deflagration @xcite , or if the wd undergoes one or more pulsations with mixing prior to the transition to detonation @xcite .
mixing may also be produced or enhanced by heating of @xmath66 rich regions , well after the thermonuclear explosion , on time scales of hours to days . both mechanism
are compatible with fluctuations on scales observed .
because of the known diversity of sne ia , below we speculate on some alternative scenarios for modeling the specific case of snr 1885 . a variation of ` classical ' dd models ( which assume _ ad hoc _ that deflagration converts into a detonation ) , are the so - called ` confined - detonation ' models recently suggested by @xcite and @xcite
. their proposal is motivated in part by the extreme sensitivity to initial conditions of the early phase of burning . in this model , a buoyancy driven individual plume rises from the central deflagrating region to the surface , where it triggers a he - detonation .
a detonation wave then propagates inward , exploding the wd .
the confined - detonation model may offer a natural mechanism to burn a wd partly via deflagration and partly by detonation .
current models of confined - detonation use only a limited nucleosynthetic network , so there is more work to be done , but the results to date suggest that burning in the central layers occurs at densities well below the value of @xmath132 @xcite needed to produce a central region with electron capture , such as is inferred from snr 1885 .
moreover , the confined detonation models predict a significant amount of fe at high velocities , which is not seen in normal sne ia .
mergers may contribute to the sne ia population and are obvious candidates for off - center explosions .
for s - andromeda , merger scenarios will suffer from the same problem as confined detonations namely , a low central density with negligable electron capture , unless the primary star is close to @xmath65 when the detonation is triggered .
however , a detonation in a @xmath133 mass wd would produce a large amount of ni and consequently an over - luminous sne ia which is incompatible with observations of the sn 1885 event .
finally , one might even consider sn 1885 as a possible core - collapse supernova .
the fact that the remnant is freely expanding out to @xmath134 indicates that the remnant has so far swept up less than about @xmath135 of material , implying an ambient circumstellar or interstellar density less than about @xmath136 .
this relatively low density , the historical record , and the absence of evidence of recent star formation in the bulge of m31 , all point to a snia event , as opposed to a snib or snic event in which a massive star undergoes core collapse after having lost its hydrogen envelope in a wind .
our acs observations add further evidence in this direction , the abundance and distribution of fe and ca being more consistent with a deflagration / detonation event than with he - rich freeze - out .
the near spherical symmetry of the image can also be construed as supporting the sn ia interpretation , since there is growing evidence that the explosion mechanism of these objects is intrinsically asymmetric with axis ratios of 2:1 or more ( @xcite ) .
the remnant of sn 1885 offers an unique and remarkable opportunity to investigate the explosion dynamics and chemistry of a sn ia explosion in some detail .
the _ hst _
acs data on snr 1885 presented in this paper provide only a glimpse of the potential to probe sn ia physics . in this light , we thought it useful to consider future observations that might provide further insights .
figure [ uvtime ] shows a model uv absorption spectrum at three different times , respectively 90 , 120 , and 140 years after the explosion .
the predicted spectrum is rich with strong , broad absorption features from dozens of resonance lines many of which evolve in strength fairly rapidly . of course
, the relative faintness of the bulge of m31 in the uv will limit the observability of the snr 1885 spectrum .
however , useful observations can be made even with limited signal - to - noise .
for example , as discussed in
[ ionization ] , the remnant is undergoing a period of rapid ionization of and to the singly - ionized state , so that uv continuum absorption by neutral ca and fe is currently decreasing rapidly .
this reduction in continuum opacity could be detected by broad - band uv imaging over a period as short as 5 years .
future spatially resolved uv spectral imaging , similar perhaps to that available with the space telescope imaging spectrograph ( _ stis _ ) , could provide a significant step forward in our understanding of sn ia events .
for example , figure [ uvtiling ] shows uv spectra taken with slits at various offsets from the remnant s center .
spectra such as these would provide a goldmine of information about the 3-dimensional structure and composition of the supernova ejecta , and hence about uncertain aspects of explosion physics in sne ia .
while narrow slits would of course provide the most precise mapping of ejecta , even broad slit spectra would provide useful information .
figure [ uvslit ] shows predicted uv spectra taken with various different slit widths available using the stis on _
sn 1885 , exploding as it did so close to , and to the near side of , the nuclear bulge of m31 , presents a remarkable opportunity to map the chemical and density structure of freely - expanding ejecta in a type ia supernova . in this paper
, we present high angular resolution _ hubble space telescope _ images of the remnant of sn 1885 sensitive to , , , and resonance line absorption .
the highest quality image is that of , which shows a spherical remnant @xmath137 in diameter , corresponding to a free - expansion radius of @xmath138 . although the image is partially saturated ( optical depth @xmath12 a few ) , it shows indications of a ring - like interior structure peaking at a radius of about @xmath43 , with a few apparent absorption ` clumps ' with an angular size near @xmath8 ( @xmath139 ) .
the sn 1885 remnant appears somewhat lopsided and smaller ( dia .
= @xmath140 ) in than , indicating that the inner parts of the remnant are more neutral than the outer parts , as expected if the ejecta are being photoionized by uv light from the bulge of m31 .
the remnant s and absorption structures appear similar in size and morphology , except that the image shows a small absorption peak of diameter @xmath8 displaced to the northeast @xmath9 ( projected velocity of @xmath141 ) from the center of the structure . finally , a weak detection of snr 1885 across the wavelength range of @xmath142 is probably caused by near saturated absorption from several resonance lines .
the image appears similar in size to that seen in .
we argue that the off - center and ring - like characteristics of the ca and images favor off - center delayed - detonation models of the explosion .
we considered alternative scenarios including pure deflagration models , confined - detonations , and wd mergers , but find all of these to be less likely .
using the hydra code @xcite , we have made detailed computations that follow delayed - detonation models of sn 1885 from the beginning of nuclear runaway up to the present 120 year age of the remnant .
the models follow in detail the nucleosynthesis in , radiative transfer through , and ionization structure of , the ejecta , including the effects of late - time photoionization by uv light from the bulge of m31 .
we selected a best - match model so as to reproduce the observed maximum velocity of ca .
the best - match model undergoes a transition from deflagration to detonation at a mass fraction of @xmath92 .
the model is subluminous by about @xmath76 magnitudes , consistent with indications from the historical record of sn 1885 .
the model synthesizes @xmath75 of @xmath66 , and has a light curve similar to that of the subluminous type ia event , sn 1986 g .
we find that our model is able to reproduce many of the general observational features of snr 1885 .
ca and fe in snr 1885 are mostly singly - ionized at the present time , so the and images trace essentially all the ca and fe in the remnant . by contrast , and trace only the regions with the highest density and abundance of ca and fe , because only in these regions are and sufficiently self - shielding against photoionizing uv from the bulge of m31 to survive up to the present time .
our analysis presents the first direct comparison between the properties of a supernova remnant and multi - dimensional explosion models to probe the underlying physics of burning fronts . within the framework of delayed - detonation ( dd ) models ,
the deflagration - detonation transition occurs at densities of the order of @xmath143 , consistent with the zeldovich mechanism @xcite .
potential candidate processes for mixing of burned and unburned matter may be instabilities internal to the deflagration front @xcite , during a pulsational phase of the wd @xcite , or externally induces by shear instabilities present in rapidly , differentially rotating wds when rising plumes enter this region of strong differential rotation @xcite .
the apparent presence of small scale -rich clumps is the first direct evidence for some instabilities and thus the existence of a deflagration phase in sne ia or , alternatively , mixing induced by radioactive decay of @xmath11ni over time scales of seconds or days .
however , the degree of mixing allowed by the observed images is much smaller than current 3d calculations for rayleigh - taylor dominated deflagration fronts .
this could indicate a serious problem in the current generation of multi - dimensional models for deflagration fronts , which may be related to the initial conditions at the time of the runaway .
in addition , the images require a central region of no or little ca but iron group elements indicative for burning under sufficiently high densities for electron capture taking place , i.e. , burning prior to a significant pre - expansion of the wd .
off - center , delayed - detonation models have the virtue of being able to reproduce light curves and spectra of normal and subluminous sne ia @xcite .
it is remarkable that these same models are able to reproduce the distribution and morphology of ca and fe in the remnant of sn 1885 .
however , it should be emphasized that the models described in this paper are parametrized , and thus not fully self - consistent .
our efforts should be regarded only as a first step toward tightening the link between explosion physics and observations of young supernova remnants .
the details of sne ia remain poorly understood @xcite . how the deflagration begins and develops is extremely sensitive to initial conditions @xcite .
equally unclear is how the deflagration subsequently turns into a detonation .
while observations of any one object are unlikely to completely resolve such difficult questions , further observations of the sn 1885 remnant might yield additional insights and guidance for future computer simulations of sn ia explosions .
this work was supported by nasa through grant go-10118 to raf , mch , and pah from the space telescope science institute , which is operated by the association of universities for research in astronomy .
pah is also supported by the nsf grant ast-0507557 .
clg is supported through uk pparc grant ppa / g / s/2003/00040 and jcw by nsf ast-0406740 .
baade , w. 1946 unpublished ; see osterbrock , d. e. 2001 , walter baade - a life in astrophysics ( princeton : princeton univ .
press ) , 9394 benz , w. , cameron , a. g. w. , press , w. h. & bowers , r. l. 1990 , , 348 , 647 boulesteix , j. , georgelin , y. p. , lecoarer , e. , marcelin , m. , & monnet , g. 1987 , , 178 , 91 brown , t. m. , ferguson , h. c. , smith , e. , kimble , r. a. , sweigart , a. v. , renzini , a. , & rich , r. m. 2004 , , 127 , 2738 burstein , d. , bertola , f. , buson , l. m. , faber , s. m. , & lauer , t. r. 1988 , , 328 , 440 calder , a. c. , plewa , t. , vladimirova , n. , brown , e. f. , lamb , d. q. , robinson , k. , & truran , j. w. 2003 , , 203 , 4509 cardelli , j. a. , clayton , g. c. , & mathis , j. s. 1989 , , 345 , 245 chevalier , r. a. , & plait , p. c. 1988 , , 331 ,
l109 colella , p. , woodward , p.r .
1984 , j.comp.phys .
54 , 174 crane , p. c. , dickel , j. h. , & cowan , j. j. 1992 , , 390 , l9 de bruyn , a. g. 1973 , , 26 , 105 de vaucouleurs , g. , & corwin , h. g. 1985 , , 295 , 287 dickel , j. r. , & dodorico , s. 1984 , , 206 , 351 dominguez i. , hflich p. 2000 , apj 528 , 854 dwarkadas , v. , chevalier , r. a. 1998 , , 497 , 807 eck , c. r. , cowan , j. j. , & branch , d. 2002 , , 573 , 306 ferguson , h. c. , & davidsen , a. f. 1993 , , 408 , 92 fesen , r. a. , gerardy , c. l. , mclin , k. m. , & hamilton , a. j. s. 1999 , , 514 , 195 fesen , r. a. , saken , j. m. , & hamilton , a. j. s. 1989 , , 341 , l55 ford , h. c. , et al .
1998 , , 3356 , 234 fransson , c. , lundqvist , p. , chevalier , r. a. 1996 , , 461 , 993 freedman , w. l. , & madore , b. f. 1990 , , 365 , 186 fryxell b. , arnett d. , mller e. 1991 , apj 367 , 619 gamezo , v. n. , khokhlov , a. m. , & oran , e. s. 2005 , , 623 , 337 gamezo , v. n. , khokhlov , a. m. , oran , e. s. , chtchelkanova , a. y. , & rosenberg , r. o. 2003 , science , 299 , 77 gaposchkin , c. p. 1936
, , 83 , 245 graham , j. r. 1988 , , 326 , l51 hamilton , a. j. s. , & fesen , r. a. 1991 , in supernovae , 10th santa cruz summer workshop in astronomy and astrophysics , ed .
s. e. woosley ( berlin : springer - verlag ) , 656 hamilton , a. j. s. , & fesen , r. a. 2000 , , 542 , 779 hflich , p. 1995
, , 443 , 89 hflich , p. 2003 , asp conf .
288 , stellar atmosphere modeling , ed .
i. van hubeny , d. mihalas and k. werner ( san francisco : asp ) , 185 hflich , p. 2006
, nuclear physics a , vol .
777 , 579 hflich , p. , wheeler , j. c. , and thielemann , f.k .
1998 , apj 495 , 617 hflich , p. , gerardy , c. , fesen , r. a , & sakai , s. 2002 , , 568 , 791 hflich , p. , gerardy , & c. , linder , e. , et al .
2003 , in stellar candles for the extragalactic distance scale , lecture notes in physics , 635 , 203 hflich , p. , gerardy , c. l. , nomoto , k. , motohara , k. , fesen , r. a. , maeda , k. , ohkubo , t. , & tominaga , n. 2004 , , 617 , 1258 hflich , p. , gerardy , c. , & quimby , r. 2004 , in cosmic explosions in three dimensions , ed .
p. hoflich , p. kumar & j. c. wheeler ( cambridge : cambridge university press ) , 142 hflich p. , khokhlov , a. , wheeler j.c .
1995 , , 444 , 211 hflich p. , & stein j. 2002 , , 568 , 771 hoyle , p. , & fowler , w. a. 1960 , , 132 , 565 iben jr , i. , & tutukov , a. v. 1984 , , 54 , 335 iwamoto , k. , brachwitz , f. , nomoto , k. , kishimoto , n. , umeda , h. , hix , w. r. , & thielemann , f. 1999 , , 125 , 439 janka , h .-
t . , scheck , l. , kifonidis , k. , mller , e. , & plewa , t. 2005 , in asp conf .
332 , the fate of the most massive stars , ed .
r. humphreys and k. stanek ( san francisco : asp ) , 363 joshi , y. c. , pandey , a. k. , narasimha , d. , sagar , r. , & giraud - hraud , y. 2003 , , 402 , 113 kaaret , p. 2002
, , 578 , 114 khokhlov , a. , mller , e. , & hflich , p. 1993
, , 270 , 223 khokhlov , a. , oran , e. s. , & wheeler j. c. 1997 , , 478 , 678 khokhlov , a. 2001 , astro - ph/0008463 kraus , j. d. , & dixon , r. s. 1965 , , 70 , 324 kurucz , r. 1995 , atomic line data ( r.l .
kurucz and b. bell ) kurucz cd - rom no .
cambridge , mass . : smithsonian astrophysical observatory leonard , d. c. , & filippenko , a. v. 2001 , , 113 , 920 liu , w. , jeffery , d. j. , & schultz , d. r. 1997 , , 483 , l107 livne , e. 1999 , , 527 , l97 livne , e. 2004 , in cosmic explosions in three dimensions , ed .
p. hoflich , p. kumar & j. c. wheeler ( cambridge : cambridge university press ) , 110 livne , e. , asida , s. m. , & hflich p. 2005
, , 632 , 443 lynas - gray , a. e. , seaton , m. j. , & storey , p. j. 1995 , journal of physics b atomic molecular physics , 28 , 2817 mcconnachie , a. w. , irwin , m. j. , ferguson , a. m. n. , ibata , r. a. , lewis , g. f. , & tanvir , n. 2005 , , 356 , 979 middleditch , j. , & kristian , j. 1984 , , 279 , 157 mihalas , d. 1978 , stellar atmospheres ( 2nd ed ; new york : freeman ) , 650 minkowski , r. 1939 , , 89 , 156 morton , d. c. 1991 , , 77 , 119 niemeyer , j. , & woosley , s. e. 1997 , , 475 , 740 & 524 , 290 paczyski , b. 1985 , in cataclysmic variables and low - mass x - ray binaries , ed . d. q. lamb and j. patterson ( dordrecht : reidel ) , 1 patchett , b. e. , stickland , d. j. , crilly , d. , & wood , r. 1985 , the observatory , 105 , 232 pavlovsky , c. , et al .
2004 , in acs instrument handbook , version 5.0 , ( baltimore : stsci ) phillips , m. m. 1993 , , 413 , l105 phillips , m. m. , et al .
1987 , , 99 , 592 phillips , m. m. , lira , p. , sunzeff , n. b. , schommer , r. a. , hamuy , m. , & maza , j. 1999 , , 118 , 1766 plewa , t. , calder , a. c. , & lamb , d. q. 2004 , , 612 , l3 rich , r. m. , corsi , c. e. , cacciari , c. , federici , l. , fusi pecci , f. , djorgovski , s. g. , & freedman , w. l. 2005 , , 129 , 2670 rpke , f. k. , niemeyer , j. c. , & hillebrandt , w. 2003 , , 588 , 952 rpke , f. k. , & hillebrandt , w. 2005 , , 429 , l29 ruiz - lapuente , p. , lucy , l. 1992 , , 400 , 127 sjouwerman , l. o. , & dickel , j. r. 2001 , aip conf .
proc . 565 , young supernova remnants , ed .
s. s. holt and u. hwang ( aip ) , 433 spencer , j. h. , & burke , b. f. 1973 , , 185 , l83 stanek , k. z. , & garnavich , p. m. 1998 , , 503 , l131 thielemann f .- k .
, nomoto k. , hashimoto m. 1994 , in : supernovae , les houcheseds s.s .
bludman r. , mochkovitch j. , mochkovitch j. , zinn - justin elsevieramsterdam , p 629 uenishi , t. , nomoto , k. , & hachisu , i. 2003 , , 595 , 1094 van den bergh , s. 1994 , , 424 , 345 van den bergh , s. 2002 , , 123 , 2045 verner d. a. , ferland , g. j , korista , k. t. , yakovlev , d. g. 1996 , , 465 , 487 wang , l. , howell , a. , hflich , p. , & wheeler , j. c. 2001 , , 550 , 1030 webbink , r. f. 1984 , , 277 , 355 whelan , j. , iben jr . , i. 1973 , , 186 , 1007 yoon , s .- c .
, langer , n. , & scheithauer , s. 2004 , , 425 , 217 yoon , s. c. & langer , n. 2004 , , 419 , 623 | sn 1885 was a probable subluminous type ia supernova which occurred in the bulge of the andromeda galaxy , m31 , at a projected location 16@xmath0 from the nucleus .
here we present and analyze _ hubble space telescope _ images of the sn 1885 remnant seen in absorption against the m31 bulge via the resonance lines of , , , and .
viewed in h & k line absorption , the remnant appears as a nearly black circular spot with an outermost angular radius of @xmath1 , implying a maximum linear radius of @xmath2 pc at m31 s estimated distance of @xmath3 kpc and hence a 120 yr average expansion velocity of @xmath4 .
the strongest absorption is organized in a broken ring structure with a radius of @xmath5 (= @xmath6 km s@xmath7 ) with several apparent absorption ` clumps ' of an angular size around that of the image pixel scale of @xmath8 (= 1500 km s@xmath7 ) . and absorption structures appear similar except for a small absorption peak displaced @xmath9 off - center of the structure by a projected velocity of about @xmath10 km s@xmath7 .
analyses of these images using off - center , delayed - detonation models suggest a low @xmath11ni production similar to the subluminous sn ia explosion of sn 1986 g .
the strongly lopsided images of of and can be understood as resulting from an aspherical chemical distribution , with the best agreement found using an off - center model viewed from an inclination of @xmath12 60@xmath13 .
the detection of small scale ca ii clumps is the first direct evidence for some instabilities and the existence of a deflagration phase in sne ia or , alternatively , mixing induced by radioactive decay of @xmath11ni over time scales of seconds or days
. however , the degree of mixing allowed by the observed images is much smaller than current 3d calculations for rayleigh - taylor dominated deflagration fronts .
moreover , the images require a central region of no or little ca but iron group elements indicative for burning under sufficiently high densities for electron capture taking place , i.e. , burning prior to a significant pre - expansion of the wd . using time - dependent ionization calculations , we show that the presence today of neutral ions in this 120 yr old remnant can be understood as ejecta self - shielding from the uv radiation in the m31 bulge . |
the casimir effect @xcite , inspiring many branches of physics @xcite , features a decisive geometry dependence : the fluctuation - induced interaction between test bodies or surfaces depends on their shape and orientation .
this is because the casimir effect arises from the fluctuation spectrum in presence of the surfaces relative to the vacuum fluctuations .
the spectral properties in turn are a direct consequence of the geometry .
this geometry dependence becomes even more pronounced at finite temperature @xmath0 : thermal fluctuations can predominantly be associated with a characteristic length scale , the thermal wavelength @xmath1 .
thermal fluctuations contribute to the casimir force , whenever the scale set by the thermal wavelength is commensurate with a mode of the fluctuation spectrum as defined by the geometry .
therefore , thermal corrections to the zero - temperature casimir effect generally can not be described by universal additive terms or other simple recipes but require a careful analysis of the interplay between geometry and temperature , as first anticipated in @xcite .
this `` geothermal '' interplay has first been verified in paradigmatic perpendicular - plates @xcite or general inclined - plates configurations @xcite .
further evidence for the experimentally relevant sphere - plate configuration has been provided recently in @xcite .
typical low - temperature dependencies in these open geometries obey power laws with characteristic exponents that are particular for the geometry .
most importantly , these power laws disagree with predictions from standard local approximation techniques such as the proximity force approximation ( pfa ) @xcite even in the limit of vanishing surface separation .
this is in contrast to zero - temperature forces which are often well described by the pfa in this limit @xcite . in this work ,
we perform a comprehensive study of the geometry - temperature interplay for the sphere - plate and cylinder - plate configuration .
we study the casimir forces induced by fluctuations of a scalar field obeying dirichlet boundary conditions on the surfaces in order to explore the geothermal interplay in a most transparent fashion .
moreover , we use the worldline approach to the casimir effect @xcite which on the one hand provides for a highly intuitive picture of the fluctuations , and on the other hand facilitates analytical as well as numerical computations from first principles @xcite .
for instance , the failure of local or additive approximation techniques can directly be inferred from the temperature dependence of the force density : the latter tends to delocalize for decreasing temperatures on scales of the thermal wavelength @xcite .
local approximation techniques may only be useful at finite temperature if the strict weak - coupling limit is taken @xcite , or in the high - temperature limit . in the present work ,
we analyze the thermal force density distributions , compute thermal forces for a wide nonperturbative range of parameters , and determine asymptotic limits .
this facilitates a careful comparison with local approximation techniques , and , most importantly , yields new and unexpected results for the geometry dependence of thermal forces . for instance , the pure thermal force , i.e. , the thermal contribution to the casimir force , reveals a non - monotonic behavior below a critical temperature for the sphere - plate and cylinder - plate case @xcite : the attractive thermal force can increase for increasing distances .
this anomalous feature is triggered by a reweighting of relevant fluctuations on the scale of the thermal wavelength a phenomenon which becomes transparent within the worldline picture of the casimir effect .
whereas these non - monotonic features already occur for a simple dirichlet scalar model , non - monotonicities can also arise from a competition between te and tm modes of electromagnetic fluctuations in configurations with side walls @xcite . while there are a number of impressive verifications of the zero - temperature casimir force @xcite , a comparison between theory and thermal force measurements suffers from the interplay between dielectric material properties and finite temperature @xcite , still being a subject of intense theoretical investigations @xcite . in view of the geothermal interplay
, we expect that the full resolution of this issue requires the comprehensive treatment of geometry , temperature and material properties , possibly also including edge effects @xcite .
first results on the sphere - plate configuration using scattering theory and specific dielectric models demonstrate this nontrivial interplay @xcite
. as a crucial ingredient for such an analysis , field - theoretical methods for casimir phenomena have to be used that can deal with arbitrary casimir geometries .
in addition to the worldline methods @xcite used in this work , a variety of approaches has been developed in recent years , such as a functional integral approach @xcite and scattering theory @xcite .
an extension of these methods to finite temperature is usually straightforward and highly worthwhile in view of the geometry - temperature interplay .
our paper is organized as follows : after a brief account of the worldline approach to the casimir effect in sect . [
sec : worldl - appr - casim ] , the sphere - plate and cylinder - plate configurations are studied at zero temperature in sect .
[ sec : case - t=0 ] .
in addition to making contact with the literature , we perform the worldline computation directly for the force instead of the interaction energy . section [ sec : fin - t ] contains all our main results on the finite - temperature case .
our conclusions are summarized in sect .
[ sec : conclusions ] . for reasons of comparison ,
the proximity - force approximation for the sphere - plate and cylinder - plate case is worked out in detail in appendix [ app : pfa ] .
in addition to explicit formulas which have not comprehensively appeared in the literature so far , we relate the pfa to an approximate treatment of the worldline path integral which helps to understand the differences between the exact and approximate treatments .
we start with a short reminder of the worldline approach to the casimir effect for a massless dirichlet scalar in 4 dimensions ; for details , see @xcite . for a configuration @xmath2 consisting of two rigid objects with surfaces @xmath3 and @xmath4 , the worldline representation of the casimir interaction energy in @xmath5 dimensional spacetime reads @xmath6 \right\rangle.\end{aligned}\ ] ] the worldline functional @xmath7 $ ] is @xmath8 if the worldline @xmath9 intersects both objects @xmath10 , and is zero otherwise . the expectation value in eq .
is taken with respect to an ensemble of @xmath11-dimensional closed worldlines with a common center of mass @xmath12 and obeying a gauian velocity distribution .
for static casimir configurations the time component cancels out at zero temperature .
equation ( [ int-1 ] ) has an intuitive interpretation : all worldlines intersecting both surfaces violate dirichlet boundary conditions and are removed from the ensemble of allowed fluctuations , contributing to the negative casimir interaction energy . during the @xmath13 integration ,
the extent of a worldline is scaled by @xmath14 .
large propertimes @xmath15 correspond to ir fluctuations , small @xmath15 to uv fluctuations .
introducing finite temperature @xmath16 by the matsubara formalism is equivalent to compactifying euclidean time on the interval @xmath17 $ ] . the casimir free energy corresponding to eq .
now becomes @xmath18 \right\rangle.\nonumber\end{aligned}\ ] ] for numerical purposes , it is convenient to remove the @xmath15 dependence from the velocity distribution by the rescaling @xmath19 where @xmath20 .
now , the @xmath21 function reads more explicitly @xmath22\equiv\theta[\mathbf x_{\text{cm}}+\sqrt{{\mathcal{t}}}{\boldsymbol{\gamma}}(t)].\ ] ] the worldline integrals are evaluated numerically by monte carlo methods , i.e , the path integral is approximated by a sum over a finite ensemble of @xmath23 worldlines .
each worldline @xmath24 is furthermore discretized by a finite set of @xmath25 points per loop ( ppl ) . to generate discretized worldlines with gauian velocity distribution
the v - loop algorithm was used in this work @xcite . in the remainder
, we apply the worldline method to the sphere - plate and cylinder - plate casimir configurations .
let us first study the sphere - plate and cylinder - plate geometries at zero temperature . here
, we make contact with earlier results for the casimir effect of a cylinder and sphere above a plate @xcite .
moreover , we generalize the worldline method to directly compute the casimir force instead of the energy which leads again to significant simplifications compared to previous energy calculations
. the method will be generalized to finite temperature in the next section .
we start the configuration of a sphere above a plate .
the sphere of radius @xmath26 is centered around the origin @xmath27 .
the infinitely extended plate lies in the @xmath28 plane , where @xmath29 is the minimal distance between both objects , see fig .
[ ceff - sap-10 ] .
since the configuration has a rotational symmetry with respect to the @xmath30 axis , the three dimensional @xmath31 integration reduces to a two dimensional one
. the casimir energy ( [ int-1 ] ) reads @xmath32 \right\rangle,\end{aligned}\ ] ] where we have switched to cylindrical coordinates ( @xmath33 ) with @xmath34 . the @xmath35 $ ] functional factorizes , @xmath36 = \theta_\mathrm{s } [ \mathbf{x}_\mathrm{cm}+\sqrt{\mathcal{t } } { \boldsymbol{\gamma}}]\theta_\mathrm{p } [ \mathbf{x}_\mathrm{cm}+\sqrt{\mathcal{t } } { \boldsymbol{\gamma } } ] .
\end{aligned}\ ] ] here @xmath37 and @xmath38 account for the intersection of a worldline @xmath39 with the sphere and the plate , respectively .
notice that @xmath37 is independent of @xmath29 , whereas @xmath38 reads @xmath40 where @xmath41 denotes the worldline s extremal extent into the negative @xmath30 direction .
as we are interested in calculating the casimir force , @xmath42 , the derivative acting only on @xmath38 produces a @xmath43 function which eliminates the @xmath44 integral .
the casimir force thus simplifies to @xmath45\right\rangle.\end{aligned}\ ] ] here , we have introduced @xmath46 the transition from worldline calculations of the force does not only lead to technical simplifications . also , the classification of relevant worldlines changes slightly : for the casimir energy in eq .
( [ int-1 ] ) the worldlines are scaled by the propertime @xmath47 with respect to their center of mass which is finally integrated over .
for a given center of mass , all points on a worldline @xmath48 lie on rays originating from the center of mass .
these rays are traced out by the @xmath15 integral running from @xmath49 to @xmath50 . plane .
the sphere of radius @xmath26 is in the origin .
the center of mass @xmath31 of the worldline is in @xmath51 . during the propertime integration the worldline always touches the plate , while all its points move on rays passing through @xmath52 . only points lying inside the cone will pass through the sphere . ]
by contrast , the casimir force in eq .
results from worldlines which are attached to the point @xmath53 on the plate . for a given point @xmath53 ,
all points on a worldline @xmath54 lie on rays which now originate from @xmath53 .
again these rays are traced out by the @xmath15 integral .
now , the plate is always touched by by construction for all values of @xmath55 , the remaining problem being the detection of intersection events with the sphere . adapting methods from @xcite , it is clear that only those points of a worldline lying on the rays intersecting the sphere eventually pass through the sphere for some values of @xmath55 .
let @xmath56 denote the set of points on such rays intersecting the sphere , with @xmath57 labeling these rays for a discretized worldline .
those values of propertime @xmath55 for which this point lies exactly on the sphere can be obtained from the equation @xmath58 equation has two solutions , @xmath59 for @xmath60 the point @xmath61 lies inside the sphere .
the point @xmath53 can be viewed as a tip of a cone that wraps around the sphere with the opening angle @xmath62 , with @xmath63 .
the value of the square root in eq .
varies between zero and @xmath26 .
the square root is zero if the ray merely touches the sphere , and @xmath26 if the ray lies on the cone s axis , i.e. , if it coincides with the direction spanned by @xmath53 . for a given @xmath64 , the worldline intersects the sphere if the propertime @xmath15 is in one of the intervals bounded by eq . for all possible values of @xmath57 .
denoting these intervals by @xmath65 $ ] .
the total support of the propertime integral then is @xmath66 the @xmath64 dependence of this support arises from the fact that the set of @xmath57 rays lying inside the cone depends on the position @xmath64 where the worldline is attached to the plate .
the casimir force ( [ ceff - sap-2 ] ) now reads @xmath67 the most time - consuming part of the algorithm is the determination of @xmath68 if the distance between the sphere and plate is small . to reduce the computational time ,
it is advisable to reduce the @xmath25 points per worldline to the subset of @xmath69 points on the above mentioned rays intersecting the sphere . for a given @xmath64 , all points on rays outside the cone
can immediately be dropped .
furthermore in the process of taking the @xmath64 integral from zero to infinity , the opening angle of the cone shrinks .
all points on rays which leave the cone through its upper half can then be dropped completely from the calculation , as they will never enter the cone again .
only rays below the cone , i.e. , between the cone and the plate , can enter the cone for larger values of @xmath64 . with these optimizations and with one integral less , the computational time for casimir force calculations
is significantly reduced compared with those of the casimir energies studied in previous worldline investigations .
these simplification facilitate to extend the previously studied parameter range to even larger @xmath70 ratios with higher statistics . in many respects ,
the cylinder - plate configuration is `` in between '' the sphere - plate configuration and the classic parallel - plates case .
this also holds for the experimental realization : the effort of keeping the cylinder parallel to the plate is less than it is the case for two parallel plates @xcite ; for the sphere - plate case , this issue is simply absent . as a clear benefit , the force can ,
in principle , be made arbitrarily large , since it is proportional to the length of the cylinder .
the geometry of the cylinder - plate configuration can be parameterized analogously to the preceding sphere - plate case : we consider the symmetry axis of a cylinder of an ( infinite ) length @xmath71 and radius @xmath26 to coincide with the @xmath72 axis .
the infinite plate lies in the @xmath73 plane , with @xmath29 being the distance between the cylinder and the plate .
the casimir force can be obtained directly from eq . , where use the fact that the @xmath74 $ ] functional factorizes ( cf .
) @xmath75 = \theta_\mathrm{cyl } [ \mathbf{x}_\mathrm{cm}+\sqrt{\mathcal{t } } { \boldsymbol{\gamma}}]\theta_\mathrm{p } [ \mathbf{x}_\mathrm{cm}+\sqrt{\mathcal{t } } { \boldsymbol{\gamma } } ] .
\end{aligned}\ ] ] here @xmath76 and @xmath38 account for the intersection of a worldline @xmath77 with the cylinder and the plate , respectively .
again , only @xmath78 depends on @xmath29 and is given in eq . .
the @xmath72 integral in the casimir energy is now trivial due to translational symmetry .
the casimir force can then be obtained directly from eq . and reads @xmath79\right\rangle,\end{aligned}\ ] ] where @xmath80 and @xmath81 as in the case of the sphere , the worldlines @xmath82 are attached to the plate at the point @xmath53 .
the only difference is that the worldlines are now @xmath83 dimensional a fact which reduces the computational cost .
only those points of a worldline lying on the rays intersecting the cylinder pass through the latter for some values of @xmath55 .
the construction of the support of the @xmath15 integral is identical to that for the sphere - plate case , such that the total casimir force on the cylinder can be written as in eq .
@xmath84 it is instructive to compare our results not only with analytic estimates , but also with the much simpler proximity force approximation ( pfa ) .
the latter is used by default for the data analysis of geometry corrections in most experiments .
it derives from a classical reasoning for generalizing the parallel - plate case ; thus , deviations of the exact result from the pfa estimate also parameterize genuine geometry - induced quantum behavior . above an infinite plate vs. the distance @xmath29 .
the force is normalized to the zeroth - order pfa formula .
we observe an excellent agreement with the exact asymptotic solutions for small @xmath29 @xcite and for large @xmath29 @xcite up to @xmath85 . for larger @xmath29
, the number of @xmath25 points per loop has to be increased far beyond @xmath86 used for this plot ; otherwise , the sphere falls through the rough mesh provided by the insufficiently discretized worldline , leading to a systematically underestimated force as is visible here for @xmath87 ( pink triangles ) . ] roughly speaking , the pfa subdivides the surfaces into small surface elements , applies the parallel - plate force or energy law to pairs of surface elements and integrates the resulting force density .
the pfa is inherently ambiguous as the measure for this final integration is not unique : possible alternatives are the surface measures of one of the involved surfaces or any intermediate auxiliary surface .
later on , we will refer to the `` sphere - based '' or `` plate - based '' pfa as two generic options for the integration measure .
the pfa for the present configuration is discussed in detail in appendix [ app : pfa ] .
the concept of the pfa can also be translated into the worldline picture : as an approximation to the ensemble of complicated multidimensional worldlines , we may reduce the worldlines to one dimensional straight lines .
the length of a line then corresponds to the average extent of a worldline into a certain relevant direction in a given geometry .
this picture also explains the occurrence of deviations from the pfa as well as the sign of these deviations in the dirichlet case : due to the spatial extent of the worldlines , they generically intersect both boundaries for smaller values of @xmath15 than simple straight lines .
as small propertimes yield quantitatively larger contributions , this property then results in a greater force .
more precisely , the size squared of a worldline is proportional to the propertime parameter which is in the denominator of the worldline formula , see eq . .
this explains why worldline results are typically underestimated by the pfa for small separations of the objects .
for very small separations , the upper bound of the propertime integration can effectively be set to infinity , whereas the lower bound is a measure for the first ( proper-)time , when a worldline intersects both objects .
above an infinite plate vs. the distance @xmath29 .
the force is normalized to the zeroth - order pfa formula .
we observe an excellent agreement with the exact asymptotic solutions for small @xmath29 @xcite and for large @xmath29 @xcite up to @xmath88 . ]
the casimir force for the sphere and cylinder is compared to the pfa estimates in fig .
[ ceff - rt0 - 1 ] and [ ceff - rt0 - 2 ] respectively . for similar comparisons for the casimir energy ,
see @xcite .
we have normalized the force to the leading - order pfa , which is exact in the limit of vanishing separation @xmath29 .
the systematic study of the pfa , also from the worldline point of view , is summarized in appendix [ app : pfa ] .
we observe that the normalized force obtained with worldline numerics does not lie inside the range spanned by the ambiguity of the pfa estimates .
most prominently , the sign of the deviations from the @xmath89 limit is different in the dirichlet scalar case , as can be understood in the worldline picture described above .
these observations have been frequently made in the literature before @xcite . in order to obtain fig .
[ ceff - rt0 - 1 ] we have used ensembles with up to @xmath90 and @xmath86 . at very small distances
@xmath29 the number of points per loop is not very important , since part of the systematic error is reduced by normalizing to the leading order result ; thus , even @xmath91 is sufficient for example for @xmath92 at a precision level of @xmath93 . on the other hand at @xmath85
the number of points per loop used was @xmath94 . for larger distances
the number of points per loop has to be increased far beyond @xmath95 , since , as shown in fig .
[ ceff - rt0 - 1 ] .
even such high resolution is not sufficient to resolve the small sphere for larger distances . already anticipating our results for finite temperature
, this observation gives us a rough estimate for the validity limits at small temperatures .
below , we observe that for @xmath96 the maximum of the thermal contribution to the force density at low temperatures @xmath97 lies outside the sphere on scales @xmath98 . from the fact that ensembles with @xmath99 are reliable for those cases where the dominant contribution to the force density lies within @xmath100
, we conclude that temperatures above @xmath101 are accessible also in the limit @xmath102 . from an algorithmic point of view
, the sphere - plate and cylinder - plate configurations differ with respect to computational efficiency also beyond the trivial dimensional factors : for a sphere at large separations , a large fraction of points of a worldline can be dropped right from the beginning , as they never see the sphere , i.e , they never lie on a ray inside the cone .
the situation is different for a cylinder .
dealing with a two dimensional problem , we use two dimensional worldlines and the number of points per worldline which now have to lie in a wedge is higher than those lying in a cone for the sphere - plate case . using comparable worldlines with a large number of points per loop , we thus expect the worldline numerics to break down at far larger distances @xmath29 than in the case of a sphere .
this is indeed the case as is visible in fig .
[ ceff - rt0 - 2 ] . for the fig .
[ ceff - rt0 - 2 ] we have also used ensembles with up to @xmath103 and @xmath86 . at @xmath85 , the number of points per loop used was @xmath104 , and increased to @xmath105 for @xmath106 and up to @xmath86 for @xmath88 . as expected , the required number of points per loop for a certain @xmath29 is less here than in the case of a sphere .
even at such large separations as @xmath88 , we observe an excellent agreement with @xcite .
the corresponding estimate for the validity limits at small temperatures then is @xmath107 .
at finite temperature @xmath16 , the free energy can be decomposed into its zero - temperature part @xmath108 and finite - temperature correction @xmath109 , @xmath110 the same relation holds for the casimir force @xmath111 . within the worldline representation of the free energy , the finite - temperature correction is purely driven by the worldlines with nonzero winding number @xmath112 .
most importantly , the complicated geometry - dependent part of the calculation remains the same for zero or finite temperature .
let us first perform a general analysis of the thermal correction for a generic casimir configuration following an argument given in @xcite .
we start from the assumption that the casimir free energy can be expanded in terms of the dimensionless product @xmath113 , @xmath114 no negative exponents should be present in eq .
, since the thermal part of the energy disappears as @xmath115 .
generically , the @xmath116 casimir energy @xmath117 diverges for surfaces approaching contact @xmath89 . from eq . , we would naively expect the same for the thermal correction .
if , however , sufficiently many of the first @xmath118 s in eq .
vanish , then the thermal part of the casimir energy is well behaved and without any divergence for @xmath119 .
this turns out to be the case for two parallel plates ( @xmath120 , and @xmath121 ) and for inclined plates ( @xmath122 , and @xmath123 ) @xcite .
consequently , an extreme simplification arises : the low - temperature limit of the thermal correction can be obtained by first taking the formal limit @xmath124 .
this was first observed in @xcite and then successfully applied in @xcite . in the following
, we argue that there is no divergence in the local thermal force density in the limit @xmath89 for general geometries . for a generic geometry ,
the @xmath29-divergent part can only arise from the regions of contact as @xmath119 .
the divergence for these regions at @xmath116 is due to the diverging propertime integral over @xmath125 which is bounded from below by @xmath126 .
this is because for worldlines smaller than @xmath29 the worldline functional is always zero . at finite temperature
the divergence in the thermal correction for @xmath119 is removed since one integrates now over @xmath127 , which is zero for every @xmath128 in the limit @xmath129 .
the only nonanalyticity could arise from the infinite sum . that this is not the case
can directly be verified : instead of integrating over the support @xmath130 , we integrate over @xmath55 from zero to infinity , yielding @xmath131 for finite temperature @xmath132 , eq . is a finite upper bound for the original local thermal force density .
this procedure corresponds to substituting the critical regions of contact by broader ( and infinitely extended ) parallel plates , see @xcite .
the thermal contribution is estimated from above by flattening the surfaces in the contact region .
the local thermal contribution to the casimir force of the original configuration is clearly smaller than the finite thermal contribution of parallel plates .
as the latter does not lead to divergences for @xmath102 , there can also be no divergence for the general curved case arising from the contact regions .
of course , infinite geometries may still experience an infinite thermal force , as it is the case for two infinitely extended parallel plates , but the local thermal contribution to the force density will be finite . from a practical viewpoint , taking the limit @xmath119 first simplifies the calculations considerably . ) for perpendicular plates ( dashed blue line ) , cylinder above plate ( dotted dashed line ) and sphere above plate ( solid red line ) both of radius @xmath133 in the zero - distance limit @xmath102 for @xmath134 .
the sphere - plate curve represents the radial density including the radial measure factor @xmath135 .
the thermal force densities of cylinder and perpendicular plates at @xmath136 are equal to the force density of two parallel plates , @xmath137 .
the thermal force density in the sphere - plate case has a maximum of @xmath138 , where the factor @xmath139 arises from the cylindrical measure .
note that a considerable fraction of the force density lies outside the sphere which only extends to @xmath140 . as the temperature drops , the maximum moves monotonously to the right .
] another important feature of low - temperature contributions to the casimir effect is the spread of the thermal force density over regions of size @xmath141 even for very small separations @xmath29 .
this phenomenon has first been demonstrated for the configuration of two perpendicular plates at a distance @xmath29 @xcite .
( in this configuration , the sphere in fig .
[ ceff - sap-10 ] is replaced by a vertical semi - infinite plate extending along the positive @xmath30 axis and an edge at @xmath142 . ) the thermal force density @xmath143 for this case as a function of the coordinate @xmath64 on the infinite surface measuring the distance from the edge ( i.e. , the contact point at @xmath124 ) can indeed be obtained analytically on the worldline from the thermal force , @xmath144 here , @xmath145 is a worldline parameter measuring the extent of half a unit worldline , i.e. , the distance measured in @xmath146 direction from the left end to the center of mass .
it is clear from fig .
[ ceff - sap-10 ] that the lower bound in the @xmath55 integral in eq .
is given by @xmath147 : this is the minimal scaling value for which the worldline intersects the semi - infinite vertical plate . from eq .
( [ ft-1c2 ] ) , we read off the following force density : @xmath148 analytic results for the thermal force can be obtained by rescaling the radial coordinate @xmath149 per worldline and using @xmath150 .
the thermal force between the perpendicular plates in the limit @xmath89 upon integration then yields @xmath151 in agreement with @xcite .
the perpendicular - plates configuration is special as it features a scale invariance in the @xmath102 limit : eq .
( [ ft-1c ] ) remains invariant under @xmath152 and @xmath153 for arbitrary @xmath154 . as a consequence , knowing ( [ ft-1c ] ) for a single temperature value , say @xmath155 , is sufficient to infer its form for all other @xmath0 .
equation ( [ ft-1c ] ) is shown for @xmath156 in fig .
[ t0-ip - fd ] .
for @xmath157 , the force density stays nearly constant , corresponding to the first term in ( [ ft-1c ] ) .
it rapidly approaches zero for @xmath158 . from this
, we draw the important conclusion that the region of constant force density in @xmath64 direction can be made arbitrarily large by choosing sufficiently low @xmath0 .
similar consequences arise for temperature effects in other geometries .
we plot the thermal force densities for the sphere - plate and cylinder - plate configuration in fig .
[ t0-ip - fd ] .
the thermal force density for a cylinder above a plate at @xmath124 has a shape similar to the one of two inclined plates , whereas the radial force density of a sphere above a plate exhibits a maximum due to the cylindric measure factor @xmath64 , see fig .
[ t0-ip - fd ] .
although these force densities are not scale invariant due to the additional dimensionful scale @xmath26 ( sphere radius ) , its maximum nevertheless moves away from the sphere as the temperature drops .
we conclude that no local approximate tools such as the pfa will be able to predict the correct thermal force in particular at low temperatures .
the fact that the force densities for sphere and cylinder are not scale invariant leads to different temperature behaviors for @xmath97 and @xmath159 even in the limit @xmath119 . and for @xmath133 .
the worldline result and the pfa predictions are normalized to the leading - order pfa , cf .
( [ ft - sap-3 ] ) .
the leading - order pfa predicts a @xmath160 behavior of the thermal force for all @xmath0 . on the other hand , for small @xmath0 the plate - based pfa and the worldline result
see a @xmath161 behavior , whereas the sphere - based pfa sees a @xmath162 one . for large @xmath163
the behavior is @xmath160 for all curves .
all predictions agree in the large @xmath0 limit .
for very small @xmath0 , the worldline result runs into the blue area which spans the pfa predictions . ]
let us start with the expansion of the thermal force for @xmath164 and for small temperature @xmath165 .
following our general argument given above , no singularities in @xmath29 appears in the limit @xmath102 . also , we expect that the thermal force decreases with decreasing @xmath26 .
this motivates an expansion of the thermal force with only positive exponents for @xmath29 and @xmath26 .
assuming integer exponents , dimensional analysis permits @xmath166 from our numerical results in the limit @xmath89 , we observe a @xmath161 behavior of the thermal force , see fig .
[ ft - sap - f1 ] .
we conclude that @xmath167 is negligible with respect to @xmath168 in the regime @xmath169 where numerical data is available .
in fact , we conjecture that vanishes identically @xmath170 ; if so , also @xmath171 vanishes , since the configuration would otherwise be more sensitive to temperatures at small @xmath29 than at @xmath124 .
our conjecture is supported by the following argument based on scaling properties : the dimensionless ratio of the thermal correction and zero - temperature force has to be invariant under the rescaling @xmath172 the same holds for the ratio of @xmath173 at @xmath124 and the zero - temperature force at @xmath174 . for @xmath164
, we can use the pfa for the zero - temperature force , which to leading order yields @xmath175 . if @xmath176 , this leading ratio would be @xmath177 which is invariant under the rescaling ( [ ft - sap-1b ] ) ; in addition , this ratio would be invariant under ( [ ft - sap-1b ] ) with @xmath26 fixed . if @xmath170 , then this ratio is @xmath178 which is invariant only under the full transformation ( [ ft - sap-1b ] ) .
the result that for @xmath179 the thermal correction would exhibit the same @xmath26 dependence as the zero - temperature force for small distances @xmath180 is counterintuitive : whereas the radial force density in the small - distance limit at @xmath116 is peaked right under the sphere near @xmath181 , the thermal correction arises from contributions at much larger @xmath64 , cf .
[ t0-ip - fd ] . as a simple estimate
, we expect that the thermal correction is proportional to an effective area of the sphere , @xmath182 , as seen by the worldlines .
this estimate then is compatible with @xmath183 and @xmath184 .
the question arises why the pfa approximation yields a @xmath160 behavior despite the additional scale @xmath26 .
the reason is that @xmath26 appears only in the combination @xmath185 in the force density , such that @xmath186 , @xmath187 leaves the force density invariant up to a multiplicative constant . for the sphere - plate configuration in the limit @xmath89 for @xmath133 . for @xmath188 ,
we observe a linear behavior which can be fitted to @xmath189 .
we have used @xmath190 loops with @xmath191 ppl each . for @xmath192 ,
the number of points per loop used is not sufficient to resolve the sphere properly , inducing systematic errors ( black triangles ) . ]
let us return to eq .
( [ ft - sap-1 ] ) . for @xmath168 and @xmath193
, we obtain numerically ( see figs .
[ ft - sap - f2 ] and [ ft - sap - f3 ] ) @xmath194 these numbers can be confirmed by the exact @xmath0-matrix representation @xcite .
note that both coefficients have the same sign , implying that the absolute value of the thermal correction to the casimir force _ increases _ with increasing @xmath29 for sufficiently small @xmath29 and @xmath0 .
this apparently anomalous behavior can be understood in geometric terms within the worldline picture @xcite .
the system has a critical temperature @xmath195 : for @xmath196 , the thermal force decreases monotonically for increasing sphere - plate separation @xmath29 in accordance with standard expectations . for smaller temperatures @xmath197 , the thermal force first increases for increasing separation , develops a maximum and
then approaches zero as @xmath198 .
the peak position is shifted to larger @xmath29 values for increasing thermal wavelength , i.e. , decreasing temperature . in all cases ,
the force remains attractive , see fig .
[ ft - sap - f4 ] . as an example
, room temperature @xmath199k corresponds to the critical temperature for spheres of radius @xmath200 m . for larger spheres ,
room temperature is above the critical temperature such that the thermal force is monotonic . for smaller spheres ,
the thermal force is non - monotonic at room temperature .
if , for instance , @xmath201k and @xmath202 m , the thermal force increases up to @xmath203 m .
the high - temperature limit @xmath163 agrees with the pfa prediction for @xmath102 and reads @xmath204 in the limit @xmath102 , the pfa yields eq .
( [ ft - sap-3 ] ) for _ all _ @xmath0 .
this is because geometrically the leading - order pfa corresponds to approximating the sphere by a paraboloid , which is a scale - invariant configuration at @xmath124 . at finite @xmath29 ,
the scale invariance is broken and a term @xmath205 appears on the right - hand side of eq .
( [ ft - sap-3 ] ) at low temperature in the leading - order pfa .
by contrast , we observe that the true @xmath89 limit is characterized by a @xmath161 behavior for small @xmath0 and @xmath160 behavior for large @xmath0 . also , the sign of the correction at finite @xmath29 is different : the full worldline result predicts an increase whereas the pfa correction reduces the absolute value of the force , see fig .
[ ft - sap - f3 ] .
-dependent part of the ( negative ) thermal casimir force for a sphere at @xmath206 and @xmath133 in the low - temperature regime . in the range @xmath207 , the worldline data for @xmath208 is well approximated by @xmath209 ( dashed blue curve ) , corresponding to a @xmath193 coefficient in eq .
( [ ft - sap-1 ] ) @xmath210 .
note that the absolute value of the casimir force has increased with @xmath29 .
we have used @xmath211 loops with @xmath191 ppl each . ]
it is interesting to compare our results to another pfa scheme beyond the leading - order pfa : the plate based pfa .
this scheme is not scale invariant at @xmath124 , as the low - temperature limit for @xmath164 is also quartic and given by @xmath212 equation , in fact , corresponds to the thermal force density of two parallel plates integrated over the area of the region below the sphere , @xmath213 .
numerically , the corresponding worldline coefficient is more than ten times larger than the pfa prefactor @xmath214 . in eq .
( [ ft - sap-4 ] ) , the low-@xmath0 behavior at finite @xmath29 is exponentially suppressed , implying that the plate based pfa prediction for @xmath193 is zero which is again in contradiction with our worldline analysis . the formulae ( [ ft - sap-3 ] ) and ( [ ft - sap-4 ] ) are derived in the appendix .
the thermal force at @xmath124 is shown together with the pfa predictions in fig .
[ ft - sap - f1 ] . and @xmath133 , normalized to the thermal force at @xmath124 .
for sufficiently small temperatures , the absolute value of the thermal force correction @xmath215 first _
increases _ with increasing @xmath29 .
for @xmath216 the small @xmath29 behavior is well described by @xmath217 .
this verifies the fit used in fig .
[ ft - sap - f3 ] with coefficient linear in @xmath218 in front of @xmath219 . from this prediction
, we would expect the curves to be monotonically decreasing for @xmath220 in the leading order .
due to higher - order terms , the @xmath221 curve still increases slightly first .
the statistical errors of the worldline calculation are of the order of the thickness of the curves . ] for the sphere - plate configuration normalized to the corresponding pfa coefficient . at large @xmath29 ,
the normalized coefficient is conjectured to approach a constant .
for small @xmath29 , the behavior is well described by @xmath222 . for @xmath87 , systematic errors similar to those of fig .
[ ceff - rt0 - 1 ] set in ( black triangles ) ; at finite temperature , these errors are even more pronounced due to a softer propertime exponent @xmath223 . ]
we now turn to the high - temperature limit , in a strict sense corresponding to @xmath224 and @xmath163 .
the second requirement is automatically fulfilled in the small - distance limit @xmath180 .
quantitatively , it turns out that the high - temperature regime is already approached for @xmath224 and @xmath165 .
a special case arises for @xmath102 , where the high - temperature limit agrees with the pfa prediction eq .
( [ ft - sap-3 ] ) in the leading order . for @xmath225 ,
the high - temperature limit is linear in @xmath0 and the total force becomes classical , i.e. , independent of @xmath226 .
this behavior is rather universal being a simple consequence of _ dimensional reduction _ in high - temperature field theories , or equivalently , of the linear high - temperature asymptotics of bosonic thermal fluctuations @xcite . in order to find the high - temperature limit ,
we perform the poisson summation of the winding sum .
the poisson summation for an appropriate function @xmath227 reads @xmath228 where @xmath229 is the fourier transform ( including a @xmath230 prefactor ) of @xmath227 . applying eq .
( [ ft - sap-5 ] ) to the winding sum , we obtain @xmath231 for finite @xmath29 , the propertime integral is bounded from below and the last term is exponentially vanishing as @xmath232 . evaluating the worldline integrals for the first two terms , we obtain @xmath233 the evaluation of @xmath234 is analogous to eq .
( [ ceff - sap-8 ] ) , @xmath235 where the support @xmath236 is the same as in the @xmath116 case , see eqs .
( [ ceff - sap-7 ] ) and ( [ ceff - sap-8 ] ) .
normalized to the zero temperature force for various temperatures and @xmath237 .
the worldline results ( symbols ) should be compared with the pfa estimate ( lines ) .
we observe that this ratio of thermal to zero - temperature force is surprisingly well described by the pfa for a wide parameter range , especially in the high - temperature regime .
this happens because both @xmath238 and @xmath239 increase with respect to the pfa with roughly the same rate , see fig .
[ ceff - rt0 - 1 ] and fig .
[ ft - sap - f5 ] . ]
the casimir force remains attractive also for high temperatures . the function @xmath238 , normalized to the pfa prediction @xmath240 is shown in fig .
[ ft - sap - f5 ] .
the function @xmath241 is monotonically increasing on @xmath242 ( similar to @xmath243 ) . at small @xmath29
, we obtain @xmath244 in analogy to the zero - temperature force , we conjecture that also @xmath241 remains monotonically increasing and finally approaches a constant for @xmath198 .
a consequence of this conjecture is that the high - temperature limit then has a simple form , @xmath224 , without any relation to @xmath26 . indeed
, demanding @xmath245 for a fixed @xmath0 , the limit @xmath198 corresponds immediately to @xmath224 , since @xmath243 itself approaches a constant .
our numerical data shown in fig . [ ft - sap - f5 ] is indeed compatible with this conjecture .
however , the large-@xmath29 limit is difficult to assess due to the onset of systematic errors for @xmath87 .
comparing fig .
[ ceff - rt0 - 1 ] and [ ft - sap - f5 ] , we notice that the zero - temperature force @xmath239 and the high temperature coefficient @xmath238 behave similarly .
this is not surprising since @xmath238 in @xmath5 minkowski space corresponds to @xmath239 in @xmath246 euclidean space due to dimensional reduction in the high - temperature limit . for finite @xmath29 , the high - temperature limit is already well reached for @xmath247 . in the pfa approximation ,
the weaker thermal force at not too small temperatures is normalized by the weaker zero - temperature force , leading to an accidental cancellation , such that for @xmath248 @xmath249 independently of @xmath26 .
a comparison between the full worldline result and the pfa for the normalized force is shown in fig .
[ ft - sap - f6 ] for various @xmath29 and @xmath0 .
since for small separations @xmath237 , the pfa is a reasonable approximation already at medium temperature @xmath250 , see fig .
[ ft - sap - f1 ] , we observe that the ratio between thermal casimir force and zero - temperature result is surprisingly well described by the pfa for quite a wide parameter range .
we stress that the pfa is inapplicable for each quantity alone .
analogous to the sphere - plate case , we start with the expansion of the thermal force at low temperature @xmath0 and for @xmath164 as in eq .
( [ ft - sap-1b ] ) .
again , we allow only for positive exponents for @xmath29 and @xmath26 . even though @xmath251 terms appear in an @xmath252 expansion at zero temperature , our numerical results at small finite temperatures , somewhat surprisingly , are consistent with an expansion of the type @xmath253 the potential leading - order terms @xmath254 and @xmath255 are expected to be zero similar to the sphere - plate case , see above , since the configuration of a cylinder above a plate is not invariant unter @xmath256 and @xmath257 .
we have no evidence for a term @xmath258 , which would lead to a nonanalytic increase of the force .
thus at small temperatures , the powers of @xmath0 are found to be integers in leading order .
similar to the sphere - plate case , we expect the low - temperature contributions to the thermal force to be proportional to the effective area @xmath259 seen by the distant worldlines .
this results in the rough estimate @xmath260 , which also implies that both coefficients have the same sign . in the limit @xmath102 .
the worldline result and the pfa predictions are normalized to the leading - order pfa eq .
( [ ft - cap-3 ] ) .
the leading - order pfa predicts a @xmath261 behavior of the thermal force for all @xmath0 . on the other hand ,
worldline numerical data is compatible with a @xmath161 behavior for small @xmath165 and a @xmath160 behavior for large @xmath159 .
this is also observed in the plate - based pfa , whereas the cylinder based pfa goes as @xmath162 for small @xmath0 .
all predictions agree in the high - temperature limit .
the worldline result enters the blue area for small @xmath0 into , which is the region spanned by the different pfa approximations . ] in the limit @xmath89 , our data in the regime @xmath169 is compatible with a @xmath262 behavior of the thermal force . for @xmath168 and @xmath193 ,
we obtain ( see figs .
[ ft - cap - f2 ] and [ ft - cap - f3 ] ) @xmath263 as in the case of the sphere , both coefficients have the same sign , i.e. , the absolute value of the thermal casimir force increases with increasing @xmath29 for sufficiently small @xmath29 and @xmath264 . for the critical temperature ,
we obtain @xmath265 . as in the case of a sphere , the thermal force decreases monotonically with increasing @xmath29 for @xmath266 ; below the critical temperature , the thermal force first increases up to a maximum and then decreases again approaching zero for @xmath198 .
the position of the maximum depends on @xmath0 and increases with inverse temperature , see fig .
[ ft - cap - f4 ] . in both cases
, however , the thermal force remains attractive .
the high - temperature limit @xmath163 agrees with the pfa prediction in the limit @xmath102 as expected , @xmath267 as for the sphere ,
the pfa predicts the same force law ( [ ft - cap-3 ] ) in the limit @xmath102 for _ all _ @xmath0 . at finite @xmath29 ,
the scale invariance is broken and a term @xmath268 appears on the right - hand side of eq .
( [ ft - cap-3 ] ) in leading - order pfa at low temperature .
by contrast , we observe different power laws for different temperatures in the limit @xmath102 : a @xmath161 behavior for small @xmath0 and @xmath261 behavior for large @xmath0 .
also , the sign of the finite-@xmath29 correction of the full result is opposite to that of the pfa , see fig .
[ ft - cap - f3 ] , all of which is reminiscent to the sphere - plate case . for a cylinder above a plate for @xmath133 in the limit @xmath102 . for @xmath269 , we observe a linear behavior . in the range @xmath270
, our data can be fitted to the form @xmath271 . for @xmath272 ,
systematic errors due to worldline discretization artifacts lead to a fast decrease of the data ( black triangles ) .
we have used @xmath273 loops with @xmath191 ppl each . ] incidentally , the beyond - leading - order pfa schemes reflect the correct behavior much better .
we observe that the cylinder - based pfa turns out to be the better approximation ( as for the sphere - based pfa in the preceding section ) .
this is opposite to the zero - temperature case . for the plate - based and cylinder - based pfa
, we obtain @xmath274 the plate - based result is equal to the thermal force of two parallel plates integrated over an area @xmath275 .
the plate - based coefficient is more than four times smaller than the worldline coefficient , whereas the leading coefficient of the cylinder - based formula becomes arbitrarily large as @xmath276 .
the formulae ( [ ft - cap-4 ] ) and ( [ ft - cap-4b ] ) are derived in the appendix . the thermal contribution to the force in the limit @xmath89
is shown together with the pfa predictions in fig .
[ ft - cap - f1 ] .
-dependent part of the ( negative ) thermal contribution to the casimir force for the cylinder - plate configuration at @xmath206 and @xmath133 in the low - temperature regime . in the range @xmath277 , the worldline data for @xmath208
is well approximated by @xmath278 .
we thus conclude that the @xmath193 coefficient in eq .
( [ ft - cap-1 ] ) is @xmath279 .
note that the absolute value of the casimir force has increased with @xmath29 .
we have used @xmath273 loops with @xmath191 ppl each . ]
let us now investigate the high - temperature limit , which can be obtained by poisson summation of the winding - number sum as in eq .
( [ ft - sap-6 ] ) .
a special case arises in the limit @xmath89 , where the high - temperature limit corresponds to the pfa prediction eq .
( [ ft - cap-3 ] ) in leading order . for @xmath280 ,
the high - temperature limit is again linear in @xmath0 and the total force `` classical '' , i.e. , independent of @xmath281 , @xmath282 as in eq .
( [ ft - sap-7 ] ) , we obtain @xmath283 where the support @xmath236 is the same as in the @xmath116 case , see eq .
( [ ceff - cap-3 ] ) .
the casimir force remains attractive also for high temperatures . the function @xmath238 , normalized to the pfa prediction and @xmath133 normalized to the thermal correction at @xmath124 .
for sufficiently small temperatures , the absolute value of the thermal force @xmath215 first _ increases _ with increasing @xmath29 . for @xmath216 ,
the small-@xmath29 behavior is well described by @xmath284 .
this verifies the fit found for fig .
[ ft - cap - f3 ] with a coefficient linear in @xmath29 in front of @xmath219 . from this form
, we would expect the curves to be monotonically decreasing for @xmath285 to leading order .
due to higher - order terms the @xmath221 curve still increases slightly at the beginning . ]
@xmath286 is shown in fig .
[ ft - cap - f5 ] .
the function @xmath287 is monotonically increasing for @xmath288 and is reminiscent to @xmath289 . at small @xmath29
, we obtain @xmath290 at large @xmath29 , we find using eq .
( [ ft - sap-11 ] ) and the analytical zero - temperature law @xcite , @xmath291 we can compare our results with those of an analytical result @xcite in the limit @xmath292 . the leading - order thermal contribution to the casimir force in this computation based on scattering theory reads @xmath293 where the integrand has been approximated to leading order in @xmath294 . here
, @xmath295 is a @xmath296 periodic sawtooth function which in the range from @xmath297 to @xmath296 is given by @xmath298 .
the authors of @xcite have given a simple estimate of the integral for the limit @xmath299 by replacing @xmath300 by @xmath301 and carrying out the resulting integral .
we compare our worldline results with eq . as well as with the simple estimate in fig .
[ ft - cap - f6 ] and [ ft - cap - f7 ] . here
, we propose another estimate which is valid for arbitrary @xmath302 . in this case , the sawtooth function is approximately constant for @xmath303 .
we approximate the logarithm by inserting the value @xmath304 for which @xmath305 is maximal : @xmath306 .
in turn for @xmath307 , the logarithm can be approximated by insertion of the value @xmath304 where @xmath308 has its first maximum : @xmath309 .
we choose the first maximum , as the integrand is oscillating for @xmath310 , such that cancellation can be expected to occur . however , choosing @xmath311 always leads to a regular @xmath312 behavior for small @xmath0 , whereas eq .
( [ ft - cap-12b ] ) changes sign at very small @xmath0 , see figs . [ ft - cap - f6 ] and [ ft - cap - f7 ] .
we thus conclude that eq .
( [ ft - cap-12b ] ) is valid for not too small @xmath0 .
normalized to the corresponding pfa coefficient for a cylinder above a plate .
the large-@xmath29 behavior is well described by eq .
( [ ft - cap-11 ] ) . at small @xmath29
, we find a behavior @xmath313 . ]
the thermal contribution to the casimir force then reads @xmath314 where @xmath315 in the emig et al .
approximation @xcite , whereas @xmath316 for @xmath302 and @xmath317 for @xmath307 in the approximation proposed here .
see figs .
[ ft - cap - f6 ] and [ ft - cap - f7 ] for the results at @xmath318 and @xmath319 respectively . in the small @xmath0 limit ,
( [ ft - cap-12c ] ) reads and @xmath133 compared with the analytic result ( `` emig et al . ''
@xcite ) and various approximations as discussed in the text . here
, we use the abbreviation @xmath320 .
remarkably , our proposed estimates using @xmath321 and @xmath322 , cf .
( [ ft - cap-12c ] ) , describe the actual behavior far better than the analytic result ( [ ft - cap-12b ] ) , which changes sign as @xmath115 . also the @xmath302 approximation using @xmath323 remains a reasonable estimate even for @xmath307 .
for the worldline data , we have used @xmath324 loops with @xmath95 ppl and @xmath325 loops with @xmath191 ppl . ] and @xmath133 compared with the analytic result ( `` emig et al . ''
@xcite ) and various approximations as discussed in the text .
even at such large separations , we observe that the analytic result ( [ ft - cap-12b ] ) becomes invalid and even changes sign as the temperature approaches zero .
incidentally our simple @xmath302 approximation using @xmath322 describes the actual behavior rather well also for smaller temperatures . for the worldline data
, we have used @xmath326 loops with @xmath327 , @xmath324 loops with @xmath328 ppl and @xmath329 loops with @xmath330 ppl . ]
@xmath331 writing this as @xmath332 , the @xmath161 coefficient @xmath333 always disappears for @xmath311 as @xmath115 . in our numerical worldline analysis ,
the systematic discretization errors lead to a vanishing of the corresponding coefficient as well , since the number of points per worldline becomes insufficient for resolution of the cylinder at very small @xmath0 . for an increasing number of points per worldline
, however , our data actually appears to point to a non - vanishing coefficient , see figs .
[ ft - cap - f6 ] and [ ft - cap - f7 ] . in any case , we expect the leading - order multipole expansion which is behind the asymptotic result to break down at low temperatures due to the geothermal interplay .
normalized to the zero - temperature force for various temperatures and @xmath237 .
the worldline results ( symbols ) are in a better agreement with the pfa estimates ( lines ) than in the case of a sphere , see fig .
[ ft - sap - f6 ] .
the normalized pfa results agree with the worldline results for not too low @xmath0 because again both @xmath238 and @xmath239 increase with respect to the pfa with roughly the same rate , see fig .
[ ceff - rt0 - 2 ] and fig .
[ ft - cap - f5 ] . ] for large @xmath0 on the other hand , eq .
( [ ft - cap-12c ] ) becomes @xmath334 we observe that the negative of the @xmath0-independent part approaches the zero - temperature limit of the casimir force for large @xmath70 faster if we choose @xmath335 rather than @xmath306 .
this choice of @xmath335 then constitutes our second estimate for @xmath302 . for not too small @xmath0
, the analytic result and the various @xmath304 approximations nicely agree with our worldline data , see figs .
[ ft - cap - f6 ] and [ ft - cap - f7 ] . for higher temperature ,
the behavior becomes @xmath336 and the different results acquire different slopes which partly disagree for @xmath232 . for @xmath337 ,
the analytic result becomes @xmath338 , the @xmath335 approximation yields @xmath339 , and the @xmath306 approximation @xmath340 .
the numerical worldline result is @xmath341 . at @xmath85 and high @xmath0 ,
the analytic result is @xmath342 , the @xmath335 approximation yields @xmath343 and the @xmath306 approximation @xmath344 .
the worldline result is @xmath345 . for large @xmath346 ,
the temperature coefficient becomes @xmath347 for both @xmath335 and @xmath348 .
for the analytic result the corresponding prefactor is greater than @xmath349 and may become @xmath350 for @xmath351 .
the corresponding worldline prefactor is @xmath352 , see eq .
( [ ft - cap-11 ] ) .
let us return to the high - temperature discussion , and finally remark that also in the case of a cylinder the thermal casimir force normalized to the zero temperature result is well described by the pfa for @xmath248 .
analogously to eq .
( [ ft - sap-12 ] ) , we conclude from the dimensional - reduction argument , that the ratio of thermal to zero - temperature force in the high - temperature limit @xmath248 is approximately @xmath353 also at medium temperatures this ratio is surprisingly well described by the pfa , even better than in the case of a sphere , see fig .
[ ft - sap - f6 ] .
the normalized thermal force is shown in fig .
[ ft - cap - f8 ] for various @xmath29 and @xmath0 .
in this work , we have analyzed the geometry - temperature interplay in the casimir effect for the case of a sphere or a cylinder above a plate . since finite - temperature contributions to the casimir effect are induced by a thermal population of the fluctuation modes , the geometry has a decisive influence on the thermal corrections as the mode spectrum follows directly from the geometry .
a strong geometry - temperature interplay can generically be expected whenever the length scale set by the thermal wavelength is comparable to typical geometry scales . within our comprehensive study of the casimir effect induced by dirichlet scalar fluctuations for the sphere - plate and cylinder - plate geometry ,
we observe several signatures of this geometry - temperature interplay : the thermal force density is delocalized at low temperatures .
this is natural as only low - lying long - wavelength modes in the spectrum can be thermally excited at low @xmath0 . as a consequence , the force density
is spread over length scales set not only by the geometry scales but also by the thermal wavelength .
this implies that local approximation techniques such as the pfa are generically inapplicable at low temperatures .
quantitatively , the low - temperature force follows a @xmath161 power law whereas the leading - order pfa correction predicts a @xmath160 behavior a result which has often been used in the analysis of experimental data . only for ratios of thermal to zero - temperature forces ,
we observe a potentially accidental agreement with the pfa prediction for larger temperatures . here , the errors introduced by the pfa for the aspect of geometry appear to cancel , whereas the thermal aspects might be included sufficiently accurately .
another signature of this geometry - temperature interplay is the occurrence of a non - monotonic behavior of the thermal contribution to the casimir force . below a critical temperature
, this thermal force first grows for increasing distance and then approaches zero only for larger distances .
this phenomenon is not related to a competition of polarization modes as in @xcite , but exists already for the dirichlet scalar case .
the phenomenon can be understood within the worldline picture of the casimir effect @xcite being triggered by a reweighting of relevant fluctuations on the scale of the thermal wavelength . from this picture , it is clear that the phenomenon is not restricted to spheres or cylinders above a plate ; we expect it to occur for general compact or semi - compact objects in front of surfaces , as long as the lateral surface extension is significantly larger than the thermal wavelength .
in fact , another consequence of the delocalized force density is that edge effects due to finite plates or surfaces will be larger for the thermal part than for the zero - temperature force .
our results have been derived for the case of a fluctuating scalar field obeying dirichlet boundary conditions on the surfaces . for different fields or boundary conditions ,
the temperature dependence can significantly differ from the quantitative results found in this work .
this is only natural as different boundary conditions can strongly modify the fluctuation spectrum . for instance , the thermal part of the free energy in the sphere - plate case exhibits different power laws for dirichlet or neumann boundary conditions in the low - temperature and small - distance limit @xcite . for future realistic studies of thermal corrections , all aspects of geometry , temperature , material properties , boundary conditions and edge effects will have to be taken into account simultaneously , as their mutual interplay inhibits a naive factorization of these phenomena .
, @xmath133 and @xmath116 compared with those of worldline numerics for a sphere and a cylinder , respectively , obtained from the @xmath116 force density .
the pfa predictions lie in the blue area which is bounded by eq .
( [ a - pfa-3a ] ) from below and by eq .
( [ a - pfa-3e ] ) from above .
the effective heights seen by worldlines are well approximated by the leading order pfa for @xmath64 not too large .
we conclude that for small @xmath70 the force is described best by the leading order order pfa , since for small @xmath70 the force density is concentrated around @xmath136 . ] , @xmath133 and @xmath116 compared with those of worldline numerics for a sphere and a cylinder , respectively , obtained from the @xmath116 force density .
note that the effective height for the cylinder is in a local maximum at @xmath136 . at greater separations
the heights seen by worldlines are on average lower than the pfa predictions resulting in greater casimir force . also at larger separations the leading order pfa
reflects best the actual situation , while the sphere / cylinder based pfa turns out to be the worst . ]
the proximity force approximation is a scheme for estimating casimir energies between two objects . in this approach ,
the surfaces of the bodies are treated as a superposition of infinitesimal parallel plates , and the casimir energy is approximated by @xmath354 here , one integrates over an auxiliary surface @xmath2 , which should be chosen appropriately .
the quantity @xmath355 denotes the energy per unit area of two parallel plates at a distance @xmath356 apart , which at zero temperature reads @xmath357 where @xmath358 for the dirichlet scalar case . as the pfa does not make any reference to boundary conditions , all the formulas in this appendix are analogously valid for the electromagnetic case ; all force formulas then have to be multiplied by a factor of two for the two polarization modes . at finite temperature , the corresponding expression is @xmath359 the distance is conventionally measured along the normal to @xmath2 .
the two extreme cases in which @xmath2 coincides with one of the two bodies provides us with a region spanning the inherently ambiguous estimates of the pfa . for a sphere at a distance @xmath29 above a plate ,
we thus integrate either over the plate ( plate based pfa ) , or over the sphere ( sphere based pfa ) .
the casimir force is then obtained by taking the derivative of ( [ a - pfa-1 ] ) with respect to @xmath29 . however , for the sphere based pfa , @xmath360 ( see below ) .
this implies that deriving the force estimate from the pfa of the energy in general is not the same as setting up the pfa directly for the force ( the latter would correspond to a surface integral over the parallel - plates force per unit area ) . in this work ,
we use the derivation via the energy ( [ a - pfa-1 ] ) .
the dependence of the pfa prediction on the choice of @xmath2 disappears in the limit @xmath119 at zero temperature to leading order .
this result shall be called leading - order pfa .
it can also be obtained by expanding the surface of the sphere / cylinder to second order from the point of minimal distance to the plate and then using the plate based pfa for this expansion .
the corresponding expressions for @xmath356 read @xmath361 for @xmath362 , we integrate over @xmath363 $ ] , for @xmath364 over all @xmath64 and for @xmath365 , @xmath366 over @xmath367 $ ] with an appropriate measure .
note that right underneath the sphere / cylinder all @xmath356 are equal to @xmath29 . demanding @xmath368 for @xmath369 we can transform the integration over @xmath370 into an integration over @xmath64 in a simple way by the substituting @xmath371 .
the integral then goes from @xmath372 to @xmath26 , and the corresponding @xmath356 reads @xmath373 also a measure factor resulting from @xmath374 and @xmath375 have to be taken into account . at zero temperature , we can absorb these factors into the new effective height @xmath376 with or without the prefactors @xmath377 is always greater than @xmath362 and @xmath364 and diverges for @xmath378 .
since the factor approaches @xmath8 for small @xmath64 , all functions @xmath356 coincide in this limit .
the pfa can also be developed within worldline formalism . calculating the casimir force density for two parallel plates
, we have to determine that value of propertimes @xmath55 for which one dimensional worldlines , attached to one of the plates , touch the other plate for the first time .
this event is encoded in the lower bound of the proper time integral , whereas the upper bound is set to infinity .
thus , we obtain @xmath379 the representation ( [ a - pfa-4 ] ) is suitable for zero and low temperatures , whereas for high temperatures one should use in ( [ a - pfa-4 ] ) the poisson resummed winding sum ( [ ft - sap-6 ] ) .
we encounter cumulants of worldline extents @xmath380 in low and high temperature limits which can be determined via the analytic expression @xcite @xmath381 let us now point out the difference between the pfa and the worldline approach . in the pfa , we always use one dimensional worldlines to determine the distance , whereas the worldline dimension in the full formalism corresponds to the dimension of the geometry . to obtain the casimir force for configurations containing one infinite plate in the worldline formalism , we integrate over this infinite plate as in the plate - based approach .
however , the integration does not stop at the end of the second body , which in the present case is a sphere or cylinder . at arbitrary large distances , there are still worldlines which see the sphere / cylinder , i.e. , we have to integrate to infinity .
we therefore expect the leading - order pfa to reflect best the exact force laws .
however , the propertime support is not the same , and thus worldlines see an effective height different from the one of the leading - order pfa , see fig .
[ a - f1 ] and [ a - f2 ] .
the shape of the effective worldline height is roughly the same for zero and high temperatures .
but at low temperature , the worldlines are reweighted .
only worldlines for large propertimes contribute considerably and thus worldlines at larger distances from the sphere become increasingly more important .
also their inner structure comes into play . using the plate based pfa
, we ignore these effects and take into account only the region below the sphere / cylinder with the same function @xmath362 ; hence , the result is expected to be too small . for the sphere , the evaluation of the leading - order pfa results in an especially simple expression , @xmath382 obviously , the relation @xmath383 remains valid also at finite temperature .
we thus obtain @xmath384 at finite @xmath0 and small @xmath113 ( @xmath385 ) , eq .
( [ a - pfa-4 ] ) yields @xmath386 for large @xmath387 ( @xmath388 ) , the expression ( [ a - pfa-2b ] ) leads directly to @xmath389 note that at @xmath124 the leading - order pfa predicts a @xmath160 behavior of the thermal force for all @xmath0 . at finite @xmath29 ,
the validity of the low - temperature limit is independent of @xmath26 .
with increasing @xmath29 the absolute value of the pfa thermal force is always reduced , irrespective of @xmath0 , quite the contrary to the full worldline results as discussed in the main text .
let us first analyze eq .
( [ a - pfa - s-4 ] ) for @xmath124 , @xmath391 equation ( [ a - pfa - s-5 ] ) distinguishes low- ( @xmath392 ) and high - temperature ( @xmath393 ) regimes .
the low - temperature regime is already well approached for @xmath394 . for higher @xmath0 ,
the thermal force is in the high - temperature regime , @xmath395 . at low temperatures
, we have a @xmath161 behavior which is given by the first term in eq .
( [ a - pfa - s-5 ] ) . for higher
@xmath0 , this @xmath161 term is canceled by the @xmath161 term with the exponential function , such that the leading behavior is given by the @xmath160 term . then , expanding eq .
( [ a - pfa - s-5 ] ) , we get a @xmath396 contribution : @xmath397 subtracting eq .
( [ a - pfa - s-6 ] ) from eq . ( [ a - pfa - s-5 ] ) and performing the poisson resummation , we obtain the full @xmath398 behavior at @xmath124 @xmath399 thus , the leading large-@xmath0 behavior at @xmath124 is @xmath400 .
let us now consider the case @xmath174 . for @xmath164 and low temperature @xmath401 , we have a @xmath161 behavior given by the first term in eq .
( [ a - pfa - s-4 ] ) .
the dependence on @xmath29 is exponentially suppressed .
this corresponds to the case of two parallel plates with an area of @xmath213 , where the dependence on @xmath29 is suppressed exponentially as well . at medium temperature , @xmath402 only the second and third term in eq .
( [ a - pfa - s-4 ] ) are exponentially suppressed and can be neglected .
the leading order can be found by expanding the remainder and considering only the converging sums . to find the subleading terms
, we again perform the poisson resummation . for medium temperature @xmath403 and @xmath164 , we then obtain @xmath404 the high - temperature limit for @xmath405 can be performed irrespective of the actual value @xmath70 by summing up the whole eq .
( [ a - pfa - s-4 ] ) .
the result reads @xmath406 note that eq .
( [ a - pfa - s-10 ] ) reduces to eq .
( [ a - pfa - s-3a ] ) for @xmath119 as it should . for @xmath29 larger than @xmath407 ,
the following temperature behavior occurs . at low temperature @xmath408 ,
a @xmath161 behavior arises from the first term in eq .
( [ a - pfa - s-4 ] ) . at higher temperatures ,
the behavior becomes rapidly linear as given by eq .
( [ a - pfa - s-9 ] ) , being valid for @xmath405 . for the sphere - based pfa ,
the thermal casimir force is given by @xmath410 where @xmath411 is given by ( [ a - pfa-3b ] ) .
for @xmath124 , we obtain the pfa approximation using the worldline language @xmath412 where @xmath413 is euler s constant .
the expansion in @xmath0 does not terminate after a few terms , so we concentrate on the two leading coefficients .
the coefficient in front of @xmath161 contains the worldline average @xmath414 . for an analytical expression
, we note that @xmath415 . for large @xmath416 , we get @xmath417 , such that we can expand the logarithm , @xmath418 where we have used eq .
( [ a - pfa-5 ] ) .
thus , the small-@xmath0 limit of eq .
( [ a - pfa - s-14 ] ) reads @xmath419 at @xmath124 , the pfa estimate @xmath420 lies above @xmath421 , see fig .
[ ft - sap - f1 ] . for
not too small @xmath0 , the worldline result lies above both these pfa predictions , but due to the logarithm in the @xmath161 coefficient the sphere - based pfa becomes larger at smaller @xmath0 , such that the worldline force enters the area spanned by the pfa prediction , see fig . [ ft - sap - f1 ] .
the high - temperature limit can be obtained by expanding eq .
( [ a - pfa - s-14 ] ) about @xmath422 .
the converging terms give the leading - order behavior .
for the subleading orders , one has to perform the poisson summation , however , the integral involved is rather complicated and may still be inflicted with artificial convergence problems . the leading - order behavior for @xmath124 and
large @xmath0 reads @xmath423 and corresponds to the leading behavior of the plate - based limit ( [ a - pfa - s-7 ] ) .
let us turn to the case of finite @xmath29 .
expanding eq .
( [ a - pfa - s-13 ] ) , we obtain the @xmath29 dependent part of the thermal force , @xmath424 the series has a form of @xmath425 , which we verified explicitly to 10th order .
assuming that this form holds to all orders , we get @xmath426 note that the first two terms agree with the @xmath161 coefficient of the plate - based formula ( [ a - pfa - s-9 ] ) ; we also see that the absolute value of the thermal force decreases with increasing @xmath29 . as eq .
( [ a - pfa - s-19 ] ) was obtained by interchanging summation and integration , we can not expect eq .
( [ a - pfa - s-19 ] ) to describe the full @xmath29 dependence for all @xmath29 and @xmath0 .
indeed at @xmath29 fixed , the thermal correction @xmath427 becomes @xmath336 as @xmath232 , which is clearly not the case for eq .
( [ a - pfa - s-19 ] ) . we can estimate the range of applicability of eq .
( [ a - pfa - s-19 ] ) as follows . at high temperature and @xmath428 ,
all pfa estimates agree . for large @xmath0 ,
the leading behavior is @xmath400 , see e.g. eq .
( [ a - pfa - s-17 ] ) . with increasing @xmath29
the force is still attractive .
demanding @xmath429 we see that that eq .
( [ a - pfa - s-19 ] ) leads to a positive thermal force for @xmath430 . on the other hand in the low - temperature regime ,
the @xmath124 contribution is given by eq .
( [ a - pfa - s-16 ] ) . taking only the leading contribution into account and demanding @xmath431
, we again obtain that the force becomes positive at @xmath432 .
these rather rough estimates demonstrate that the validity range for @xmath29 becomes narrower with increasing temperature .
for very small @xmath29 , however the thermal correction is linear in @xmath29 irrespectively of @xmath0 , whereas the dependence on @xmath29 in the plate - based pfa is exponentially suppressed for small @xmath433 .
unfortunately , a simple relation similar to @xmath383 does not hold any longer for the cylinder , such that the resulting formulas are not related to the known results of parallel plates and are rather complicated . for arbitrary @xmath29 and @xmath0
, we obtain where @xmath439 is the hypergeometric function in the standard notation .
( [ a - pfa - s-23 ] ) does not distinguish between @xmath237 and @xmath440 , since the relevant parameter for different temperature regions is @xmath387 .
for small @xmath113 ( @xmath441 ) , we can expand eq .
( [ a - pfa - s-23 ] ) , resulting in @xmath442 for large @xmath387 , the poisson resummation of eq .
( [ a - pfa - s-23 ] ) leads to @xmath443 which is , of course , @xmath444 .
note that at @xmath124 the leading - order pfa predicts a @xmath261 behavior of the thermal force for all @xmath0 . at finite @xmath29 ,
the validity of the low - temperature limit is independent of @xmath26 . with increasing @xmath29 , the absolute value of the thermal force
is always reduced , irrespective of @xmath0 , quite the contrary to the full worldline results . here
, we give only the analytic expressions for special limits , since no general expression could be found in a closed form . at @xmath124 and @xmath165
, the thermal force can be found from the result of two parallel plates with an area of @xmath445 , @xmath446 as temperature rises , the @xmath0 behavior changes from @xmath161 to @xmath261 . for @xmath447 and @xmath124
, the plate - based pfa agrees with the leading - order pfa , and the thermal force is given by the first term in eq .
( [ a - pfa - s-24 ] ) . at low temperatures and @xmath164 ,
the dependence on @xmath29 is exponentially suppressed , just as in the case of the plate - based pfa for the sphere .
finally , at finite @xmath29 and @xmath224 , the force becomes classical @xmath448 , with @xmath449}{480 a^{7/2 } ( 2+a)^{7/2}},\notag\end{aligned}\ ] ] and @xmath450 \zeta(3)}{8 a^{5/2 } ( 2+a)^{5/2 } \pi } .\label{a - pfa - s-28}\end{aligned}\ ] ] in eqs .
( [ a - pfa - s-27 ] ) and ( [ a - pfa - s-28 ] ) , we set @xmath133 ; general expressions can be reconstructed by simple dimensional analysis . where @xmath454 is the modified bessel function of the first kind , @xmath455 the generalized hypergeometric function , @xmath456 and @xmath457 . at small @xmath0 , the expansion of eq .
( [ a - pfa - s-29 ] ) leads to @xmath458 as temperature rises , the @xmath0 behavior changes to @xmath261 . for @xmath447 and @xmath124
, the cylinder - based pfa agrees with the leading - order pfa and the thermal force is given by the first term in eq .
( [ a - pfa - s-24 ] ) . for
sufficiently small @xmath29 the difference to the @xmath124 result reads @xmath459 at finite @xmath29 and @xmath224 , the force becomes classical @xmath460 , with @xmath461}{480 a^{7/2 } ( 2+a)^{7/2}},\notag\end{aligned}\ ] ] and @xmath462 \zeta(3)}{8 a^{5/2 } \pi ( a+2 r)^{5/2}}. \notag\end{aligned}\ ] ] in eqs .
( [ a - pfa - s-32 ] ) and ( [ a - pfa - s-33 ] ) , we set @xmath133 , general expressions can be reconstructed by dimensional analysis .
we thank t. emig for providing the data of fig .
[ ceff - rt0 - 2 ] and m. bordag and e. elizalde for useful discussions .
we have benefited from activities within esf research network casimir .
aw acknowledges support by the landesgraduiertenfrderung baden - wrttemberg , by the heidelberg graduate school of fundamental physics .
hg was supported by the dfg under contract gi 328/3 - 2 and gi 328/5 - 1 ( heisenberg program ) .
m. bordag , u. mohideen and v. m. mostepanenko , phys .
rept . * 353 * , 1 ( 2001 ) ; r. onofrio , new j. phys .
* 8 * , 237 ( 2006 ) [ arxiv : hep - ph/0612234 ] ; s. y. buhmann and d. g. welsch , prog .
quant . electron .
* 31 * , 51 ( 2007 ) [ arxiv : quant - ph/0608118 ]
. m. b. halpern and w. siegel , phys .
d * 16 * , 2486 ( 1977 ) ; a. m. polyakov , `` gauge fields and strings , '' harwood , chur ( 1987 ) z. bern and d.a .
kosower , nucl .
* b362 * , 389 ( 1991 ) ; * b379 * , 451 ( 1992 ) m.j .
strassler , nucl . phys . *
b385 * , 145 ( 1992 ) .
s. k. lamoreaux , phys .
* 78 * , 5 ( 1997 ) ; u. mohideen and a. roy , phys .
lett . * 81 * , 4549 ( 1998 ) ; h.b .
chan _ et al .
_ , science 291 , 1941 ( 2001 ) ; r.s .
decca _ et al .
_ , phys .
d * 68 * , 116003 ( 2003 ) ; phys .
lett . * 94 * , 240401 ( 2005 ) .
i. klich , j. feinberg , a. mann and m. revzen , phys .
d * 62 * , 045017 ( 2000 ) [ arxiv : hep - th/0001019 ] . v. v. nesterenko , g. lambiase and g. scarpetta , phys .
d * 64 * , 025013 ( 2001 ) [ arxiv : hep - th/0006121 ] .
m. bordag , v. v. nesterenko and i. g. pirozhenko , arxiv : hep - th/0107024 . | we investigate the nontrivial interplay between geometry and temperature in the casimir effect for the sphere - plate and cylinder - plate configurations . at low temperature ,
the thermal contribution to the casimir force is dominated by this interplay , implying that standard approximation techniques such as the pfa are inapplicable even in the limit of small surface separation .
thermal fluctuations on scales of the thermal wavelength lead to a delocalization of the thermal force density at low temperatures . as a consequence ,
the temperature dependence strongly differs from naive expectations .
most prominently , thermal forces can develop non - monotonic behavior below a critical temperature .
we perform a comprehensive study of such geothermal phenomena in these casimir geometries , using analytical and numerical worldline techniques for dirichlet scalar fluctuations . |
understanding the transport of angular momentum in accretion disks is one of the basic challenges in modern astrophysics .
the traditional approach ( @xcite ) is to assume that accretion disks are characterized by an effective viscosity , arising from an unspecified collective process , given by @xmath3 , where @xmath4 is the local sound speed , @xmath5 is the disk half - thickness , and @xmath6 is a constant of order unity .
more recently , there has been the realization ( @xcite ) that a previously discovered magnetic field instability in a shearing flow ( @xcite , @xcite ) will act to produce a positive angular momentum flux in an accretion disk .
this has given rise to two separate , but related claims .
the first is the proposal that this is the dominant mechanism of angular momentum transport in ionized accretion disks .
the second is the proposal that this instability , by itself , leads to a turbulent dynamo which drives the magnetic field into equipartition with the ambient pressure , i.e. @xmath7 , where @xmath8 is the alfvn speed in the disk . the growth rate for this dynamo is usually taken , following the original claim of balbus and hawley , to be @xmath9 .
since the dimensionless ` viscosity ' , @xmath6 , is @xmath10 , this proposal requires that @xmath6 be a number ` of order unity ' .
however , we need to interpret this phrase generously . in numerical simulations
( e.g. @xcite ) @xmath6 is less than @xmath11 both because the magnetic pressure saturates at a fraction of the gas pressure , and because the off - diagonal components of @xmath12 are a fraction of @xmath13 .
three dimensional simulations of the balbus - hawley instability have been performed by a variety of researchers , with and without imposed vertical magnetic flux , and with and without vertical disk structure ( @xcite , @xcite , @xcite , and @xcite ) .
we note in particular brandenburg et al .
( 1995 ) in which no net flux was imposed on the computational box , and vertical disk structure was included . in this simulation , as in the others , there was an initial rise in the magnetic energy density at a rate @xmath9 . at the end of this phase
the system had not yet lost memory of its initial conditions , but after a somewhat longer time , which may be as long as a few dozen rotation periods , the simulation asymptotically approached a final state with @xmath7 .
the approach to this state was characterized by the appearance of a large scale field which underwent spontaneous reversals at irregular intervals of tens of rotational periods .
interestingly , the presence of a large scale coherent field does not seem to be due to an @xmath14 dynamo , because the relative helicity is just a few percent .
conventional @xmath14 dynamo models rely on a nonzero @xmath15 component in the helicity tensor ( not to be confused with the dimensionless viscosity , written here as @xmath6 ) to produce a large scale coherent field .
the presence of an initial rapid rise is less surprising , since imposing a uniform large scale magnetic field in a turbulent medium results in the formation of intermittent magnetic field structures and a consequent rise in the magnetic energy density at the turbulent eddy turn - over rate .
in addition , there is evidence ( @xcite ) that turbulence in a conducting fluid can generate a modest and highly disordered magnetic field even in the absence of an imposed global field .
both of these effects are probably due to the ability of of symmetric turbulence to produce a negative effective diffusion coefficient ( @xcite ) and they both can be relied open to contribute to the growth of the high wavenumber component of the magnetic field . on the other hand , the slower relaxation rate seen after the initial rise
is correlated with changes in the large scale field and is presumably an indicator of the large scale dynamo growth rate .
since the turbulence is sustained by an instability of the large scale field , its ability to generate such a field is critically important .
the saturation level of the magnetic field in these simulations also leads to some puzzling questions .
the claim that the balbus - hawley instability saturates when the magnetic pressure is comparable to the ambient thermal pressure , and that the dimensionless viscosity has some approximately fixed value , is difficult to reconcile with attempts to model accretion disks in compact binary systems .
successful models of dwarf novae outbursts and x - ray transients ( @xcite , @xcite , @xcite , @xcite , @xcite , and , more recently , @xcite ) , as well as the distribution of light in quiescent dwarf novae disks ( @xcite ) all imply that the dimensionless viscosity , @xmath6 , varies spatially and with time .
these variations are consistent with @xmath16 , where @xmath17 is a constant lying somewhere between @xmath18 and @xmath19 .
recent work ( @xcite ) on x - ray transients suggests that @xmath17 may be close to @xmath20 .
here we note only that any value of @xmath17 appreciably different from zero conflicts with claims for a universal value of @xmath6 .
this difficulty can be resolved in several different ways .
for example , we might claim that magnetic instabilities dominate @xmath6 only at low temperatures and that some other process , e.g. convection , dominates at higher temperatures .
this idea faces two major objections .
first , it explains only some of the phenomenological evidence favoring a varying @xmath6 .
second , attempts to model the vertical structure of dwarf novae disks invariably conclude that such disks are convective during quiescence , when @xmath6 is small and stably stratified during outburst , when @xmath6 is relatively large ( for a recent discussion of the conditions necessary for convection in partially ionized accretion disks see cannizzo 1992 ) .
this implies that convection could explain the rise in @xmath6 only if it acts to suppress angular momentum transport , rather than enhance it .
alternatively , one could appeal to the temperature dependence of the resistivity to account for this effect , although the effective resistivity of the simulations is , in any case , many orders of magnitude larger than in real disks .
a more promising notion is that one might ascribe the rise in @xmath6 to the greater thermal conductivity of disk in the hot state , although the rationale for this is not yet clear . finally , one might simply conclude that all the phenomenological models are wrong , for a variety of reasons , a viewpoint which is difficult to dismiss given the large uncertainties faced in modeling accretion disks . in this paper we will explore a new turbulent disk dynamo in which the turbulence is not assumed to lack mirror symmetry in the vertical direction .
the dynamo effect arises from the fact that in a system of finite size the mean square helicity , and the instantaneous spatially averaged helicity , is still nonzero .
we will see that this leads to the existence of a modified @xmath14 dynamo , in which the large scale organization of the magnetic field comes from the existence of a large scale shear .
of course , a real disk has vertical structure , which breaks the vertical symmetry and allows for the _ possibility _ of a nonzero average helicity .
however , we will show that the incoherent dynamo mechanism will be particularly effective in simulations of limited azimuthal extent . indeed the simulations of brandenburg et al .
( 1995 ) show that while the relative helicity is small ( less than a few percent ) , there is still a dynamo effect leading to the generation of large scale fields . moreover
, the saturation level of the magnetic field , and the consequent value of @xmath6 , turn out to depend sensitively on the ratio @xmath21 . if this dynamo is the only effect to arise from the turbulence induced by the balbus - hawley effect , then it is relatively simple to reconcile the dynamo activity seen in simulations with phenomenological models of accretion disks in compact binary systems . in 2
we discuss the conceptual basis of an incoherent dynamo in a turbulent shearing medium and estimate the growth rate . in
3 we apply this to accretion disks and show that the incoherent dynamo gives a positive growth rate only for axisymmetric magnetic domains .
we estimate the saturated state of the field and discuss our results in light of numerical simulations of magnetic fields in a keplerian disk . in
4 we summarize our results and their implications for astrophysical disks and numerical simulations of such disks .
in a highly conducting medium the magnetic field obeys the induction equation @xmath22 where we have neglected ohmic diffusion .
ultimately this term is important in allowing reconnection and smoothing .
here we assume that these processes take place at a rate determined by turbulent processes .
the usual approach to dynamo theory is to define the response of the large scale magnetic field to small scale motions as @xmath23 and to derive its effects on the large scale field by substituting @xmath23 back into the right hand side of equation ( [ eq : ind ] ) . for an incompressible fluid this yields @xmath24 where @xmath25 is the levi - civita tensor and @xmath26 and @xmath27 the first term comes from the stretching of large scale field lines by the local turbulence .
the tensor @xmath28 describes the twisting of large scale field lines into a spiral shape .
reconnection between adjacent spirals produces a large scale field component at right angles to the original field line provided that either the degree of twisting or the large scale magnetic field strength varies in the third direction .
the second term is the usual diffusion term , modified by the presence of the third term .
we can see from equation ( [ eq : alpha ] ) that each component of @xmath29 has either a factor of @xmath30 or @xmath31 . if the local velocity field is mirror symmetric , in the sense that its statistical properties are unchanged under the transformation @xmath32 , then the time and space averaged value of @xmath29 vanishes .
this poses a significant but , as we will see , not insurmountable , obstacle to a successful dynamo .
another problem is that equations ( [ eq : dyn1 ] ) , ( [ eq : alpha ] ) and ( [ eq : diff ] ) are usually defined kinematically , i.e. the velocity field is assumed to be imposed on the magnetic field .
once the magnetic field becomes sufficiently powerful it will modify the flow , which is usually taken into account by including a correction term proportional to @xmath33 .
however , in a keplerian shearing flow the magnetic field will be unstable and the resulting turbulence will be directly correlated with the magnetic field .
nevertheless , as long as we define @xmath34 in terms of the motion of the magnetic field lines equation ( [ eq : dyn1 ] ) will remain valid , if difficult to solve . here
we will define our results in terms of the properties of @xmath29 and @xmath35 regardless of their ultimate source . in a keplerian disk
the dynamo equations can be simplified as @xmath36 and @xmath37 where @xmath38 is the rotation frequency , @xmath39 is the buoyant velocity of the magnetic field lines relative to the surrounding fluid , and @xmath40 and @xmath41 are the radial and azimuthal components of the magnetic field .
equations ( [ eq : br ] ) and ( [ eq : bt ] ) differ from equation ( [ eq : dyn1 ] ) in that we have allowed for the presence of global shearing , and magnetic field line buoyancy .
in addition , we have assumed that the diffusion matrix is diagonal , and dropped the effects of helicity on the evolution of @xmath41 , given that the shearing of @xmath40 should dominate such effects .
also , we have retained only the @xmath15 term in equation ( [ eq : br ] ) since the critical feedback term in the dynamo equations involves generating radial magnetic flux from the azimuthal component of the field .
finally , given that we are interested in applying these equations to accretion disks whose thickness is a small fraction of their radius , we have assumed that vertical gradients will dominate over radial gradients .
now let s assume that the turbulence is symmetric under @xmath32 so that @xmath42 .
although this eliminates any coherent helicity , the value of @xmath43 can still increase in a random walk . ignoring diffusion and buoyancy
we see that the formal solution for @xmath40 is @xmath44 now by hypothesis , @xmath15 is uncorrelated over time scales greater than some eddy correlation time @xmath45 . if the radial magnetic field is undergoing a random walk , then it will usually be far enough away from zero that it will not change sign every eddy correlation time .
since @xmath40 drives @xmath41 through coherent shearing , this implies that the correlation time for @xmath40 and @xmath41 is much greater than @xmath45 .
consequently , we can consider the integrand in equation ( [ eq : tvar ] ) as consisting of a rapidly varying factor , @xmath15 , multiplying a slowly varying function .
multiplying equation ( [ eq : tvar ] ) times equation ( [ eq : br ] ) and ignoring diffusion and buoyancy , as before , we see that the integral in equation ( [ eq : tvar ] ) is correlated with @xmath15 only over the last eddy correlation time @xmath45 .
consequently , we can replace the integral in the product with @xmath46 .
this implies @xmath47 where @xmath48 is the number of independent turbulent eddies in a magnetic domain , @xmath49 is the vertical wavenumber of the magnetic domain , and @xmath50 is the mean square helicity associated with a single eddy . in general this will be of order @xmath51 where @xmath52 is the root mean square turbulent velocity .
since @xmath40 is being driven incoherently we can expect it to undergo frequent reversals . in between such reversals
the shearing of the field will drive @xmath13 sharply upward . from equations ( [ eq : bt ] ) and ( [ eq : ran1 ] ) we see that the correlation time of the radial magnetic field and the growth time of the magnetic field are comparable and given by @xmath53 we note that @xmath54 has to be greater than @xmath45 in order for this estimate to be internally self - consistent , i.e. the magnetic field must be correlated over longer times than the turbulence itself . since a field reversal in @xmath41 requires that @xmath40 not only reverse its sign , but maintain it long enough to push @xmath41 through zero , it is clear that the correlation time for @xmath41 may be somewhat larger than the correlation time for @xmath40 .
we will return to this point later . by itself
this argument does not show that a succession of random twists in a shearing background can drive an exponential increase in the magnetic field .
we need to show that the growth experienced between field reversals dominates over the abrupt cancellation of the field as @xmath40 reverses itself .
we also need to show that our estimate of the growth rate given in equation ( [ eq : gr1 ] ) will dominate over turbulent diffusion for some range of magnetic domain sizes .
we can test the assertion that a series of random changes in @xmath40 can drive a dynamo by constructing a simple toy model of the process , which ignores the spatial structure of the field , but includes its dynamical evolution . assuming that @xmath15 has a stochastic component and ignoring buoyancy we can rewrite equations ( [ eq : br ] ) and ( [ eq : bt ] ) as @xmath55 and @xmath56 where @xmath57 is a stochastic variable with a correlation time @xmath45 and @xmath58 is the coherent component of @xmath59 . here
we have subsumed spatial derivatives into the definitions of @xmath60 and @xmath61 and ignored the @xmath62 term which would normally appear in the mean - field dynamo equations .
we have also assumed that turbulent damping is the same for each component of the magnetic field , which is not generally true , but simplifies the analysis without losing any essential physics .
equations ( [ eq : tbr ] ) and ( [ eq : tbt ] ) can be rewritten in a more convenient form by defining @xmath63 .
then @xmath64 and @xmath65 the magnetic field will grow exponentially if @xmath66 is negative and @xmath67 .
we can find @xmath66 by solving equation ( [ eq : stoch ] ) in terms of an unnormalized probability distribution function @xmath68 and evaluating @xmath69 the distribution function @xmath68 satisfies the equation @xmath70 or @xmath71 where we have taken advantage of the unnormalized nature of @xmath68 to set the constant of integration to @xmath72 . equation ( [ eq : distev ] ) can be solved to yield @xmath73 \int_a^{\infty}\exp\left[{-\omega r^3 + 2\alpha_{coh}r\over 2\langle \eta^2\tau_{eddy}\rangle}\right]dr .
\label{eq : pdist}\ ] ] consequently , @xmath74 \over\int_{-\infty}^\infty dy\int_y^\infty ds\exp[y^3-s^3-\gamma(y - s ) ] } , \label{eq : expa}\ ] ] where @xmath75 equation ( [ eq : expa ] ) can be rewritten by defining new variables @xmath76 and @xmath77 and integrating over @xmath78 .
we obtain @xmath79dx\over \int_0^\infty x^{-1/2}\exp[x(\gamma-{x^2\over4})]dx}. \label{eq : exact}\ ] ] when @xmath80 is small we can expand @xmath81 and obtain @xmath82 which implies that @xmath83 in other words , the magnetic field will grow exponentially roughly as fast as the estimate given in equation ( [ eq : gr1 ] ) .
this will be suppressed by turbulent diffusion only when the damping rate due to diffusion is comparable to the growth rate .
the existence of an incoherent dynamo emerges from the fact that the distribution function @xmath68 given in equation ( [ eq : pdist ] ) is biased towards negative values of @xmath84 .
this bias comes , paradoxically enough , from the coherent , positive definite term in equation ( [ eq : stoch ] ) .
when @xmath84 is sufficiently positive it evolves deterministically through @xmath85 into negative values .
( actually , @xmath40 does nt change during this phase .
this deterministic trajectory is merely a field reversal for @xmath41 . )
the end result is that whenever @xmath84 becomes large and positive it rapidly switches to being large and negative .
ultimately , the sign of the bias is determined by the sign of @xmath86 .
the frequency of such field reversals is given by examining the probability distribution at large @xmath84 when the evolution of @xmath68 is deterministic .
if we define @xmath87 as the time it takes for the field to move from some large positive value of @xmath84 to @xmath88 then from equation ( [ eq : stoch ] ) we see that @xmath89 where we have neglected @xmath58 since for @xmath84 sufficiently large its effects can be ignored .
the field reversal rate is just the limit of this rate times the statistical weight of the distribution between @xmath84 and @xmath90 . in other words ,
@xmath91 substituting equations ( [ eq : pdist ] ) and ( [ eq : taua ] ) into this result , and making the change of variables , as before , to @xmath92 and @xmath78 we have @xmath93dw dx\over \int_{-\infty}^\infty\int_0^\infty \exp\left [ w\left(\gamma-{1\over4}(w^2 + 3x^2)\right)\right]dw dx}.\ ] ] both the numerator and the denominator can be simplified by integrating over @xmath92 to obtain @xmath94dw}. \label{eq : trev}\ ] ] when @xmath80 is small this becomes @xmath95 i.e. a rate which is roughly half the e - folding rate for the magnetic field energy . when @xmath58 is large and positive we can evaluate the integrals in equation ( [ eq : exact ] ) by expanding around the maximum of @xmath96 .
we obtain @xmath97 so that @xmath98 which is the expected result for a coherent @xmath14 dynamo . in this limit the magnetic field reversal rate becomes @xmath99.\ ] ] as expected , field reversals are exponentially suppressed as we go to the usual @xmath14 dynamo . given @xmath100 then as @xmath58 becomes significant we expect it to enhance the dynamo growth rate and reduce the rate of spontaneous field reversals .
when @xmath58 is large and negative we can evaluate equation ( [ eq : exact ] ) by integrating the denominator by parts and remembering that the bulk of the contribution to the integral comes from @xmath101 so that @xmath102 .
we obtain @xmath103 and @xmath104 in this limit field reversals occur at a rate given by @xmath105 we note that in this case the coherent component of the helicity does not completely shut off the incoherent dynamo , even though by itself it is incapable of driving a dynamo .
instead we find that as @xmath106 increases past one the dynamo growth rate decreases inversely with @xmath106 .
eventually , turbulent diffusion will suppress the dynamo .
in the limit where @xmath80 is of order @xmath107 we anticipate that the dynamo growth rate will be less than expected from the incoherent dynamo alone and the rate of field reversals will be larger .
these analytic results have the advantage of being based on a solvable model , but do not include the effects of spatial structure or saturation .
it is therefore instructive to consider the combined effects of a random electromotive force and shear in a one dimensional model .
we consider the mean field equations for a uniform disk with keplerian rotation and half - thickness @xmath108 , @xmath109 @xmath110 with @xmath111 and @xmath112 at @xmath113 .
we first consider the incoherent @xmath114-effect , so we take @xmath114 to be random in space and time .
when the rms value of @xmath114 is large enough we find self - excited solutions that grow without bound . in reality
there must be some quenching mechanism , which we model using @xmath115 where @xmath116 is a random function in space and time with zero mean and an rms value of unity . without loss of generality we put @xmath117 . in fig .
1 we plot contours of the @xmath41 field in a space - time diagram for a dynamo number @xmath118 of @xmath119 .
( the critical dynamo number for dynamo action depends on the coherence time and length scales , @xmath120 and @xmath121 , respectively . in the present case
we adopt @xmath122 and @xmath123 and find the critical dynamo number to be around 2000 . at this dynamo number
the ratio of the growth rate given in equation ( [ eq : gr1 ] ) to @xmath124 is @xmath125 . )
the remarkable result is that the @xmath41 field shows a great deal of spatio - temporal coherence with variations comparable to the diffusion time and diffusion length .
experiments with different dynamo numbers suggest that the degree of coherence is more pronounced for larger dynamo numbers . in order to isolate the effect of a spatially incoherent @xmath114-effect we now investigate a model with a steady @xmath114-effect of the form @xmath126 for large values of @xmath17 , the critical dynamo number is proportional to @xmath127 .
thus , although the rms value of the @xmath114-effect is unchanged , the dynamo becomes harder to excite if @xmath114 is chopped into many domains of different sign .
the magnetic field is steady , and the radial component is of alternating sign .
however , more surprisingly , the toroidal magnetic field has the same sign for all values of @xmath128 , see fig .
this is very similar to the simulation of a random incoherent @xmath114-effect mentioned before .
there is one difference in that the magnetic field shows global reversals in time when the @xmath114-effect is incoherent in time .
finally , we note that in order for the magnetic field to grow the growth rate given in equation ( [ eq : gr1 ] ) has to be greater than the dissipation rate . in general
the dissipation rate will depend on the wavenumber of the magnetic domain as @xmath129 , while the growth rate goes as @xmath130 . clearly whether or not there is a self - excited dynamo will depend in large part on the geometry of the fluid .
a keplerian accretion disk with a root mean square alfvn speed of @xmath8 will be subject a local instability first described by velikhov ( 1959 ) .
its pivotal role in transporting angular momentum outward in accretion disks was recognized later ( @xcite ) . in the context of accretion disks
this instability is normally referred to as the balbus - hawley instability .
its maximum growth rate is of order @xmath131 , and occurs at an azimuthal wavelength of @xmath132 . in three dimensions the instability saturates in turbulence with a typical turbulent velocity comparable to @xmath8 and a typical eddy size of @xmath132 .
this turbulence is not expected to be isotropic , but the typical eddies are expected to have axis ratios of order unity , which in this context means only that no axis should be more than an order of magnitude larger than another ( @xcite ) . numerical simulations ( @xcite ) indicate that the the azimuthal scale of the typical eddies is several times the vertical and radial scales , which is expected in light of the large local shear .
the azimuthal velocity is also larger , although only by a factor of roughly two .
neglecting such factors , these scaling laws imply a turbulent diffusivity of @xmath133 .
the turbulence largely suppresses the parker instability and the typical buoyant velocity of the magnetic field is of order @xmath134 ( @xcite , @xcite ) , where @xmath4 is the local sound speed . the angular momentum flux induced by the turbulence is approximately @xmath135 which implies a dimensionless viscosity @xmath6 of order @xmath136 . since @xmath137 this implies that magnetic flux is lost from the disk at a rate which is some fraction of order unity times @xmath138 .
it is by no means obvious that in real disks this turbulence possesses the kind of symmetry that would make @xmath42 . on the other hand , calculations done without vertical structure or
any imposed large scale field ( @xcite ) give results which are qualitatively similar to calculations which include vertical structure ( @xcite ) . by construction
the former calculations must be symmetric under the transformation @xmath32 even though the latter are not .
we can in principle estimate @xmath15 using data from the simulation of brandenburg et al .
( 1995 ) . from equation ( [ eq : alpha ] )
it is clear that a time integration has to be carried out .
however , video animations of those data suggest that the life time of turbulent eddies is shorter than the life time of magnetic structures which , in turn , is shorter than the eddy turn over time . in other words ,
the strouhal number ( e.g. krause & rdler 1980 ) is small . as a rough approximation we may therefore replace the time integration by a multiplication with a relevant time scale .
we adopt the natural time scale @xmath139 , which is sufficient since we are only interested in relative variations .
we adopt volume averages and note that because of the periodic boundary conditions in the toroidal direction , @xmath140 , so we can compute @xmath141 in fig .
3 we plot the evolution of @xmath15 using the data from run c of brandenburg et al .
( 1995 ) , which has now been carried out for an additional 200 orbits , see also torkelsson et al .
this average was computed for the upper half plane of the simulation .
we note that @xmath15 is positive , in agreement with the expected effect for bubbles that expand as they rise in a keplerian disk . however , the sign of @xmath15 suggested by the correlation between the azimuthal magnetic and electric fields is negative ( @xcite ) .
the source of this discrepancy is not yet clear . in any case
the spatially averaged helicity shows large variations from its long term average , although the variations in the electromotive force are much larger .
the size of the fluctuations in the electromotive force , as well as the persistence of the dynamo in the absence of any @xmath142 symmetry breaking , implies that any preferred helicity resulting from vertical structure is not strong enough to completely dominate the simulations .
in what follows we will assume that real disks lack any significant @xmath143 .
at a minimum our results can be taken as demonstrating that there is an incoherent dynamo operating in the simulations , and in real accretion disks , whose effects need to be understood , and cleanly separated from any other dynamo mechanisms at work .
let s consider a magnetic domain characterized by the wavenumbers @xmath144 . ignoring the anisotropies in the turbulence
we find that the number of turbulent eddies per domain is roughly @xmath145 .
consequently the growth rate for the dynamo is @xmath146 however , in a shearing environment we are nt free to specify @xmath147 and @xmath148 separately .
the shear implies a minimum @xmath147 for any @xmath148 since in a time @xmath149 the shear will increase @xmath147 by an amount @xmath150 .
if we choose a value of @xmath147 above this minimal value then @xmath151 will go as @xmath152 while the dissipation rate scales as @xmath153 .
clearly our chances for a successful dynamo will be maximized by taking @xmath154 .
this gives us @xmath155 this analysis only makes sense in the limit where the magnetic domains encompass at least one eddy , or @xmath156 and @xmath157 .
the dissipation rate is roughly @xmath158 by comparing equations ( [ eq : nax1 ] ) and ( [ eq : diss ] ) we see that the incoherent dynamo is incapable of generating non - axisymmetric large scale magnetic fields .
the dissipation rate of such domains exceeds the generation rate for all domain sizes greater than a single eddy . in a real disk
the number of eddies in a magnetic domain does not increase indefinitely as @xmath159 .
the finite circumference of the disk implies that for axisymmetric domains @xmath160 consequently , we can rewrite equation([eq : grge ] ) as @xmath161 at a fixed wavenumber , and therefore at a fixed dissipation rate , this rate is maximized for @xmath162 . assuming this ratio we see that the dynamo growth rate for axisymmetric domains goes as @xmath163 , which implies that at some sufficiently small @xmath164 the dynamo will work .
more exactly , the incoherent dynamo caused by the balbus - hawley instability will drive an increase in the magnetic field strength if @xmath165 in other words , the incoherent dynamo only works for @xmath166 ultimately @xmath49 is limited by the height of the disk , i.e. @xmath167 . moreover , as we approach this limit the buoyant loss of magnetic flux becomes significant .
the buoyant loss rate from a single magnetic domain goes as @xmath168 so when @xmath169 buoyant losses are as important as turbulent diffusion . of course , the only limit on the radial extent of a magnetic domain is @xmath170 , but lowering @xmath147 past @xmath171 will lower the growth rate without affecting the dissipation rate . from equation ( [ eq : blim ] )
we see that the magnetic field associated with scales of order the disk thickness will be the strongest and will be given by @xmath172 this in turn implies that the dimensionless viscosity associated with this dynamo mechanism is @xmath173 we expect this scaling law to hold only in the limit @xmath174 . as @xmath175 corrections of order @xmath176
will become important in our formula for buoyancy . since the saturation limit for the magnetic field involves the small difference between the growth rate dependence on @xmath8 , which has an exponent of @xmath177 , and the buoyant loss rate dependence , which goes as @xmath178
, we expect the saturation strength of the magnetic field to be extremely sensitive to such corrections unless @xmath179 .
we also note that the disk radius enters into this result only through its role as the circumference of an annulus .
computer simulations typically involve a short arc in place of a full annulus . in this case
the azimuthal length of the simulation has to be used in place of @xmath1 in equation ( [ eq : alpha1 ] ) .
the rate of spontaneous magnetic reversals expected in the absence of any coherent component to @xmath15 is comparable to dynamo growth rate .
however , while current simulations seem to show a significant reversal rate in the presence of vertical structure ( @xcite ) , in its absence the field can evolve for 100 orbital times without reversing ( @xcite ) . the exact relationship between the dynamo growth rate and the field reversal rate dependent on the particular model for the process , and the zero dimensional model used in this paper
may well overestimate the rate of spontaneous field reversals . nevertheless , such reversals are an intrinsic part of the model , and should occur if the simulation is run for several growth times .
the sharp rise in the field reversal rate when vertical structure is included suggests that a significant coherent @xmath15 is present in the such simulations .
( in the zero dimensional model this would argue for a negative vertical gradient in the coherent @xmath15 .
the situation is less clear in three spatial dimensions . ) in order for this helicity to allow the buildup of a coherent field in these simulations , as well as in accretion disks , it has to scale with the local rms turbulent velocity more steeply than the square of the incoherent dynamo growth rate , or @xmath180 .
if this helicity is due to the parker instability and if , as has been argued elsewhere ( @xcite ) , the balbus - hawley instability reduces the parker instability to vertical motions of order @xmath181 , then we can estimate the magnitude of the helicity as @xmath182 where @xmath183 is the correlation time for these buoyant motions . since shearing imposes the requirement that @xmath184 and since these motions are approximately incompressible
, i.e. @xmath185 , this gives a helicity less than @xmath186 if the coherent helicity has this dependence , then it becomes important only as the dynamo saturates due to turbulent mixing and buoyancy . in this case
it will not suppress the incoherent dynamo in simulations with smaller @xmath187 , or in real accretion disks , but it will remain significant in the saturated state .
the fact that the buoyancy does not significantly enhance the loss of magnetic flux is a critical element in the derivation of equation ( [ eq : alpha1 ] ) .
consequently , environments that increase magnetic buoyancy will saturate at much lower field strengths .
as an example we can consider magnetic flux tubes in a radiation pressure dominated environment . in this case we have ( @xcite ) @xmath188 combining this result with equation ( [ eq : gra1 ] ) for @xmath189 yields @xmath190 the exponent given in equation ( [ eq : alpha2 ] ) is perhaps a bit large in comparison to the value suggested by the phenomenology of disks , but this is a considerably less serious problem than if it were too small .
competing dynamo mechanisms and/or hydrodynamic angular momentum transport mechanisms could be driving @xmath6 up . how does this compare to other sources of viscosity in disks ?
the result given in equation ( [ eq : alpha2 ] ) has an extremely uncertain coefficient .
current numerical simulations give @xmath6 of order @xmath11 or less , which would suggest that this coefficient is very small . on the other hand ,
these simulations have @xmath7 and are definitely not in the asymptotic regime where our scaling laws should be valid .
we have already noted that the small difference in the exponent of @xmath8 in the dynamo growth rate and the dissipation rate , coupled to the presence of corrections to both these rates of order @xmath176 , makes it difficult to extrapolate from current results .
if the saturation value of @xmath191 approaches its asymptotic dependence on @xmath21 gradually as @xmath192 then the final value of the coefficient will be much larger than @xmath11 .
bearing in mind the large role that numerical viscosity plays in the simulations ( @xcite ) it seems prudent to regard the coefficient as an unknown numerical constant . on the other hand , since @xmath174 for many realistic disks we can compare this dynamo mechanism to others based purely on the value of the exponent in the scaling relationship .
of course , a purely local mechanism _
coherent _ mechanism will not scale with @xmath187 at all , although it might show some dependence on the local disk temperature .
however , as we noted earlier this model seems to conflict with phenomenological studies of dwarf novae and x - ray transients .
internal waves , excited by tidal instabilities in binary system disks ( @xcite ) will produce an effective @xmath6 which scales as @xmath193 ( @xcite ) .
this will be a competing mechanism for angular momentum transport in gas pressure dominated disks , and potentially the dominant one in radiation pressure dominated disks .
( although , such conditions are most likely in agn disks , where the potential for the tidal excitation of waves is less certain . )
given the nonlocal nature of the angular momentum transport mediated by internal waves , the existence of a purely local mechanism might be important , even if it does not clearly dominate .
when the disk is ionized and when internal waves are present , then the waves are capable of driving a dynamo with a growth rate @xmath194 ( @xcite ) and perhaps faster , depending on the nature of the turbulent cascade of wave energy ( @xcite )
. the resulting value of @xmath6 will be @xmath195 ( @xcite , @xcite ) .
when these conditions are met this would appear to be a more important dynamo mechanism , although once again we note that nonlocal effects on the wave - driven dynamo make the two processes somewhat incommensurate .
an equivalent estimate based on purely local physics was given by meyer & meyer - hoffmeister ( 1983 ) .
however , this estimate is based on using large scale buoyant cells driven by magnetic buoyancy , a picture which is inconsistent with turbulence in the disk ( @xcite , @xcite ) . in addition , they assumed approximate isotropy of the helicity tensor and offered a calculation of @xmath196 instead of @xmath15 .
this assumption of isotropy is inconsistent with the notion that the motions are driven by magnetic buoyancy , which for @xmath174 will have a time scale much longer than the local shearing time scale .
finally , we note that our result for the incoherent dynamo in an accretion disk is sensitive to our assumption that the process must be self - exciting .
given an external source of fluid turbulence , e.g. convection or the turbulent cascade of energy set off by finite - amplitude internal waves , the incoherent dynamo may have a significantly larger growth rate and give rise to a larger @xmath6 .
in this paper we have shown that mean - field dynamo theory allows for the existence of a new kind of @xmath14 dynamo , which we have named the incoherent @xmath14 dynamo , in which there is no coherent helicity whatsoever . in this class of dynamos the magnetic field is driven by a combination of a random walk for @xmath40 and its shearing , which creates @xmath41 .
the resultant large scale field derives its organization from coherent shearing effects , rather than any loss of mirror symmetry in the turbulence .
although this kind of dynamo necessarily includes spontaneous field reversals , such reversals may occur at a rate which is some fraction of the dynamo growth rate .
the existence of a mean - field dynamo in a flow with a mean helicity of zero is interesting for its own sake , since it provides an example of how large scale order in the magnetic field can arise from the interaction between a large scale shear and statistically symmetric local motions . in this sense
it represents an alternative to models which seek to explain dynamo activity through asymmetric turbulence and a coherent helicity .
it differs from previous attempts to do without a coherent helicity ( e.g. montgomery et al .
( 1984 ) and gilbert et al .
( 1988 ) ) in that it does so without appealing to the other terms in equation ( [ eq : dyn1 ] ) .
this model is particular interesting in light of previous claims ( @xcite ) that the coherent @xmath114 effect does not converge ( although see kraichnan ( 1979 ) for a counterargument ) .
this dynamo is particularly interesting in light of simulations of magnetic field instabilities in accretion disks .
we have suggested that this dynamo can operate successfully in accretion disks , but only to produce axisymmetric large scale fields . comparing the growth rate for this dynamo with the buoyant loss rate for magnetic flux
, we see that _ if _ this is the only dynamo associated with magnetic shearing instabilities , then the large scale magnetic field will saturate when @xmath197 and @xmath198 .
this result may seem somewhat odd , since the dynamic equations do not depend on @xmath1 at all .
however , the factor of @xmath1 comes in through geometrical considerations , i.e. from considering the number of independent eddies in an axisymmetric magnetic field domain . in that sense
it refers to the circumference of such an annulus rather than its radius .
consequently , when comparing numerical simulations to the predictions of this model one should substitute the azimuthal extent of the simulations for @xmath1 . for current simulations
this gives @xmath199 .
in fact , since the turbulent eddies are longer in the azimuthal direction , the number of independent eddies that can be stacked end to end in current simulations is one or two , so that effectively @xmath200 .
a critical test of this model would be to run simulations where this ratio was small and look for a drop in @xmath191 in the saturated state .
it s interesting to note that the only numerical simulation with no imposed field and with disk vertical structure does seem to have a coherent component to the helicity , which may have the wrong sign to drive a conventional dynamo . this may account for the large rate of spontaneous field reversals when vertical structure is put in the simulations .
the existence of such a component is consistent with the existence of an incoherent @xmath14 dynamo , but only if its amplitude scales steeply with the strength of turbulence in the disk .
finally , we note that this model successfully reconciles phenomenological models of stellar accretion disks and the existence of a dynamo effect in a magnetized disk .
the only drawback is that this model gives a relationship between @xmath6 and @xmath187 which is probably too steep , implying the existence of other , more efficient dynamo mechanisms in accretion disks in binary systems .
this work has been supported in part by nasa grant nag5 - 2773 .
in addition , etv would like to thank the harvard - smithsonian center for astrophysics for their hospitality while this paper was being written . | we use the mean - field dynamo equations to show that an incoherent alpha effect in mirror - symmetric turbulence in a shearing flow can generate a large scale , coherent magnetic field .
we illustrate this effect with simulations of a few simple systems . in accretion disks ,
this process can lead to axisymmetric magnetic domains whose radial and vertical dimensions will be comparable to the disk height
. this process may be responsible for observations of dynamo activity seen in simulations of dynamo - generated turbulence involving , for example , the balbus - hawley instability . in this case
the magnetic field strength will saturate at @xmath0 times the ambient pressure in real accretion disks .
the resultant dimensionless viscosity will be of the same order .
in numerical simulations the azimuthal extent of the simulated annulus should be substituted for @xmath1 .
we compare the predictions of this model to numerical simulations previously reported by brandenburg et al .
( 1995 ) . in a radiation pressure
dominated environment this estimate for viscosity should be reduced by a factor of @xmath2 due to magnetic buoyancy . |
recent measurements of luminosity distances using supernovae type ia ( snia ) @xcite , of the cosmic microwave background ( cmb ) temperature anisotropies with the wilkinson microwave anisotropy ( wmap ) satellite @xcite , large scale structure @xcite , the integrated sachs
wolfe effect @xcite , and weak lensing @xcite , strongly suggest that the universe is currently undergoing a phase of accelerated expansion -see
@xcite for recent reviews . within general relativity
the obvious candidate to explain the present acceleration is the cosmological constant ( or vacuum energy ) , and in fact the nave model built on it , @xmath3cdm , seems to pass reasonably well all cosmological tests .
however , it suffers from two serious drawbacks from the theoretical side : the unnatural low value of the corresponding energy density , 123 magnitude orders larger than observed , and the so - called coincidence problem " , namely , why are the densities of matter and vacuum of the same order precisely today ? " , that requires the vacuum energy density to be 96 orders of magnitude smaller than the matter density at the planck scale .
( it is fair , however , to mention the existence of proposals in which a vacuum energy density of about the right order of magnitude stems from the casimir effect at cosmic scales -see @xcite and references therein ) .
this is why much attention has been devoted to models featuring an evolving and nearly un - clustered form of energy , usually dubbed dark energy " , possessing a strong negative pressure high enough to drive late acceleration -see
@xcite for an ample review of models . for simplicity
, most cosmological models assume that matter and dark energy interact gravitationally only . in the absence of an underlying symmetry that would suppress a matter - dark energy coupling ( or interaction )
there is no a priori reason to dismiss it .
further , the coupling is not only likely but inevitable @xcite and its introduction is not more arbitrary than assuming it to vanish . on the other hand , it may help explain the coincidence problem . ultimately , this question will have to be resolved observationally . among other things ,
the interaction can push the beginning of the era of accelerated expansion to higher redshifts and it may erroneously suggest ( if the interaction is ignored when interpreting the data ) an equation of state for the dark energy of phantom type -see @xcite and references therein .
another question , is the form of the coupling .
there is no clear consensus on this point and different versions , that arise from a variety of motivations , coexist in the literature .
cosmological models where dark matter ( dm ) and dark energy ( de ) do not evolve separately but interact with each other were first introduced to justify the small current value of the cosmological constant @xcite and nowadays there is a growing body of literature on the subject -see , e.g. @xcite and references therein .
recently , various proposals at the fundamental level , including field lagrangians , have been advanced to account for the coupling @xcite .
lagrangians of scalar field coupled to matter generically do not generate scaling solutions with a dark matter dominated period lasting long enough as required by cosmic structure formation @xcite . in this paper
we compare the interacting quintessence model ( iqm ) of chimento _ et al . _
@xcite ( see also its forerunner @xcite ) with observational data ( supernovae , cosmic microwave background ( cmb ) , and matter power spectrum ) to set limits on the strength of the interaction dm / de .
the model was built to simultaneously account for the late phase of acceleration in the framework of einstein relativity and significantly alleviates the coincidence problem .
it evades the limits set in @xcite and is compatible with a right succession of cosmic eras -radiation , dark matter , and dark energy dominated expansions . in a recent paper , guo _
et al . _ also set constraints on interacting quintessence models @xcite .
however , the interactions studied by these authors differ from the one in @xcite , and while they use the cosmic background shift parameter and baryon acoustic oscillations alongside supernovae data to constrain the interaction , they do not consider the matter power spectrum , whereby our analysis may be viewed as complementary to theirs .
the outline of this paper is as follows .
next section studies the critical points of the autonomous system of equations associated to the iqm @xcite .
section iii considers the restrictions set by amendola _
et al . _ on the model to conclude that the latter evades these restrictions and , in particular , that an early epoch of baryon dominance is possible only if the strength of interaction is unnaturally large ( beyond the limits set by the cmb data ) .
section iv focus on the observational bounds coming from the cmb , matter power spectrum and recent supernovae type ia data .
finally , section v summarizes our results .
if the quintessence de decays into dm , both energy densities evolve differently than in non - interacting quintessence cosmologies and , therefore , the interaction can be tested by its effects on the dynamical evolution of the universe . due to the interaction ,
the fraction of cold dark matter ( cdm ) at any time in the past is smaller than in non - interacting models with the same cosmological parameters today .
since the dark matter energy density grows more slowly the beginning of the period of accelerated expansion and the temporal evolution of the gravitational potential differ from non - interacting models .
observables such as the angular and luminosity distances depend on the time evolution of the energy density . but the effect does not only occurs at zeroth order ; the evolution of first order matter density perturbations is also affected and so is the pattern of anisotropies of the cmb .
this section describes the dynamical evolution of dark matter and dark energy densities at zeroth order in the interacting quintessence model of ref .
@xcite to point out the main differences with respect to models with no interaction . the phenomenological model of chimento _ et al .
_ @xcite ( see also @xcite ) assumes that the dark energy decays into cold dark matter thereby both energy densities do not longer evolve separately . in @xcite
the following ansatz was adopted + @xmath4 + where @xmath5 denotes the energy densities ( subscript @xmath6 for cold dark matter and subscript @xmath7 for dark energy ) , @xmath8 is the equation of state parameter of dark energy , @xmath9 the hubble function and @xmath10 the scale factor of the flat friedmann - robertson - walker metric .
the model assumes that the dark energy decays just into cold dark matter and not into any other component such as neutrinos , baryons , or photons .
decay into neutrinos have been recently studied in the literature @xcite .
the coupling with baryons is constrained by measurements of local gravity @xcite .
the above ansatz is essentially phenomenological ; it is the simplest interaction model that leads to a constant ratio at early and late times which certainly alleviates the coincidence problem . while at present a rigorous derivation from first principles
does not exist , the ansatz may be roughly justified as follows : the right hand side of both equations in ( [ cont ] ) must be functions , say @xmath11 and @xmath12 , of the energy densities multiplied by a quantity with units of inverse of time . for the latter
the obvious choice is the hubble factor @xmath13 , so we have that @xmath14 . by power law expanding @xmath11 and retaining just the first term we get @xmath15 . to facilitate comparison of the model with observation
it is suitable to eliminate one the two @xmath16 parameters .
thus we set @xmath17 and arrive to eq .
( [ cont ] ) . the simpler choice @xmath18 would not yield a constant dark matter to dark energy ratio at late times .
the dimensionless term @xmath19 measures to what extent the decay rate differs from the expansion rate of the universe .
thus , the model is characterized by this single parameter , @xmath20 , which also gauges the intensity of the interaction .
the lower @xmath21 , the closer the evolution of the universe to a non - interacting model is .
equations in ( [ cont ] ) , the continuity equations for photons and baryons and friedmann equation @xmath22 , constitute a closed system . in fig .
[ fig0 ] we plot the evolution of the different matter density components : cold dark matter ( thick solid line ) , dark energy ( thin solid ) , photons ( thin dot - dashed ) and baryons ( thick dot - dashed ) for two values of the interaction parameter : left and right panels correspond to @xmath23 and @xmath24 , respectively . in the figure
we can see that eqs .
( [ cont ] ) have two asymptotic solutions with @xmath25 constant : at early times and in the immediate future .
it is easy to prove that , in these two cases , @xmath26 and the constant ratio is @xmath27 and @xmath28 .
these results represent a clear conceptual advantage with respect to the @xmath3cdm model .
if the present acceleration of the universe is generated by a cosmological constant , then the initial values of dark matter and cosmological constant have to be tuned by 96 orders of magnitude at planck time , i.e. , @xmath29 . in the model described above this ratio
is fixed by the interaction rate , @xmath21 .
the initial condition problem is significantly alleviated .
for example , if @xmath30 the initial dark matter to dark energy ratio will be @xmath31 . to study the evolution of the corresponding autonomous system formed by the continuity equations of all energy components and friedmann s equation we introduce the following set of variables , + @xmath32 + with @xmath33 .
then , the dynamical equations can be recast as + @xmath34x \ , , \vspace*{.2cm}\\ y^{\prime}&=&{\displaystyle\frac}{1}{2}\left[3w_{x}x^2+u^2 + 3c^2{\displaystyle}\left(1+{\displaystyle\frac}{x^2}{y^2}\right)\right]y \ , , \vspace*{.2cm}\\ z^{\prime}&=&{\displaystyle\frac}{1}{2}\left[3w_{x}x^2+u^2\right]z \ , , \vspace*{.2cm}\\ u^{\prime}&=&{\displaystyle\frac}{1}{2}\left[3w_{x}x^2+u^2 - 1\right]u\ , , \vspace*{.2cm}\\ 1 & = & x^2+y^2+z^2+u^2 \ , , \vspace*{.2cm}\\ \end{array}\ ] ] + where a prime denotes derivative with respect to @xmath35 . as said above
, the motivation for interacting quintessence models is to solve or , at least , ameliorate the coincidence problem .
a solution will be achieved if the system ( [ contnew ] ) presents scaling solutions @xcite .
as already discussed , scaling solutions are characterized by a constant dark matter to dark energy ratio .
even more important are those scaling solutions that are also an attractor . in this way ,
the coincidence problem gets substantially alleviated because , regardless of the initial conditions , the system evolves toward a final state where the ratio of dark matter to dark energy stays constant .
.location of the critical points of the autonomous system of eqs .
( [ contnew ] ) , their stability and dynamical behavior of the universe at those points . [ cols="<,<,<,<",options="header " , ] the matter power spectrum has been recently derived from two different galaxy catalogs : 2-degree field red galaxy survey ( 2dfrgs ) @xcite , that comprises a sample of more than @xmath36 galaxies with measured redshifts , and the luminous red galaxy sample ( lrgs ) of the sloan digital sky survey ( sdss ) @xcite .
a larger study , using the sdss data release 5 , and comprising more than half a million galaxies showed similar results @xcite . in @xcite , the power spectrum from the sdss was computed producing uncorrelated minimum - variance measurements in 20 @xmath37-bands of the clustering power and its anisotropy due to redshift - distortions , with narrow and well - behaved windows functions in the range @xmath38 @xmath39 .
we used these data to fit the theoretical power spectrum , @xmath40 , calculated with a modified cmbfast software and thus constrain the parameters of the cosmological model .
table [ tab : priors1 ] shows the priors imposed on the parameters in our mcmcs runs .
in addition , we allowed the bias parameter to vary in the range @xmath41 .
we run 4 different mcmcs till the criteria of convergence and well mixing were met .
figure [ fig : sdss_c2 ] shows the joint confidence contours of @xmath21 with other representative parameters . like in fig .
[ fig : c2wmap3 ] , contours run almost parallel to the vertical axis . in table
[ exp - values ] we give the cosmological parameters measured from this data set .
we derived an upper limit of @xmath42 at @xmath43 confidence level .
this constraint is weaker than the one provided by wmap 3rd year data , but applies to a different range in redshift .
this bound sets an upper limit on the value of @xmath21 averaged over the whole length of the matter dominated epoch .
varying @xmath44 or the total matter content changes the matter radiation equality and shifts the maximum of @xmath40 to larger scales , but it is around the maximum where the data have less statistical power .
matter and baryons change the amplitude and location of baryon acoustic oscillations but useful constraints can only be obtained by separating the physics of the oscillations from that governing the overall shape of the power spectrum @xcite loosing the information on the interaction . as a result
, we obtained no useful constraints on cosmological parameters except on @xmath21 .
we also run a set of mcmcs for the 2dfgrs data , but while the values of the main cosmological parameters were in very good agreement with those of table [ exp - values ] , the constraints were even poorer than those obtained from the sdss data , so we do not present them here . at low redshifts ,
only the dark matter and dark energy densities contribute substantially to the dynamical evolution of the universe .
since the interaction alters their relative ratio , it affects the actual distance to objects .
distance measurements , like the luminosity distance , @xmath45 , obtained from standard candles like the snia , are sensitive to the dark energy parameters , @xmath46 , @xmath47 , and the strength of the interaction , @xmath21 .
the likelihood function is given by @xcite , + @xmath48 + in this expression , @xmath49 is the distance modulus , @xmath50 is the uncertainty on the supernova redshift ( in units of the distance modulus ) due to its intrinsic motion and the peculiar velocities of the parent galaxy and @xmath51 is the error in the determination of the distance moduli due to photometric uncertainties .
subindexes @xmath52 and @xmath53 indicate theoretical and observed uncertainties , respectively . since the intrinsic luminosity of snia is not known , the distance moduli are insensitive to @xmath44 , only 3 parameters ( @xmath54 , @xmath8 , and @xmath21 ) are needed to characterize a model .
we carried out a brute force analysis sampling homogeneously the parameter space : ( @xmath55 , @xmath56 , @xmath57 ) ; each interval was subdivided in 10 steps .
figure [ fig : likesn ] shows the confidence intervals for @xmath21 vs @xmath46 after marginalizing over @xmath47 .
we remark that , in this case , the data is almost insensitive to @xmath21 , as we could have expected from fig [ fig : cmb3rd](c ) . to solve the coincidence problem a decay of de into dm was postulated and , consequently , a new parameter , @xmath20 , was introduced to gauge the strength of the coupling . in this subsection
we explore whether the existence of this coupling ( i.e. , a non - vanishing @xmath20 value ) it is also justified by a statistical description of the data .
the introduction of further parameters will lead to a better fit to the data .
this must be balanced against the loss of predictive power of the theory .
the bayesian information criteria ( defined as @xmath58 @xcite , where @xmath59 , @xmath37 the number of free parameters , and @xmath60 the number of data points ) , penalizes the inclusion of additional parameters to describe data of small size . for the wmap 3yr data @xmath61 and @xmath62 , so the interacting quintessence model for @xmath63 yields a slightly better fit to data than the corresponding non - interacting model .
however , the application of the bayesian criterion gives @xmath64 and @xmath65 , so the introduction of this parameter is not statistically encouraged .
this results disagree with our analysis of wmap 1yr data @xcite .
it was the unexpected consequence of our choice of priors : we fixed all models to agree with the cobe / dmr normalization .
since wmap 1yr concordance model preferred a lower normalization , it was heavily penalized compared with interacting models .
this discrepancy shows the importance of priors in bayesian analysis . as argued in @xcite ,
model selection is not a substitute for parameter fitting ; a lower bic does not imply the data rejects interacting models , just makes them unlikely .
however , they still have a theoretical advantage since they can help explain the coincidence problem , while non - interacting models can not .
let us briefly point out that while the iqm model does not solve the coincidence problem in full ( as it can not predict the present ratio of dark matter to dark energy ) it substantially alleviates it .
indeed , if we take initial conditions at the planck scale , in a model without interaction the ratio @xmath66 at that epoch is about @xmath67 . if @xmath68 , a value compatible with wmap upper limit then , asymptotically , @xmath69 , where @xmath70 is the present time .
initially @xmath71 , so , while in a @xmath3cdm model the initial conditions have to be fine - tuned by @xmath72 magnitude orders to get the present accelerated expansion , in the iqm the difference in orders of magnitude is just @xmath73 .
a further advantage is that the model is not limited by the results of ref .
@xcite since it successfully predicts that the universe undergoes three successive cosmic eras of expansion , namely , radiation , matter and dark energy dominance . if the interaction parameter were @xmath74 or larger , it could undergo an era of baryon dominance , but this is not a stable solution and , in any case , @xmath20 values larger than @xmath75 do not appear compatible with cmb data .
the existence of a late accelerated expansion of the universe requires @xmath76 , as in noninteracting models . whether or not the present value of the ratio @xmath66 coincides with the final attractor value depends on the parameters @xmath21 and @xmath8 .
these are unknowns that can be constrained by observation .
our analysis shows that the cmb , matter power spectrum and luminosity distance data only set upper limits on the dm / de coupling and , statistically , the introduction of this extra parameter is not favored by the data . at present
, cmb data sets the tightest upper limit on the strength of the interaction ( @xmath77 ) . as fig .
[ fig : excluded ] reveals , this upper bound can be translated into an excluded region ( shaded area ) for the dark matter to dark energy ratio , @xmath66 .
solid lines , from top to bottom , correspond to @xmath78 and @xmath79 that bounds the shaded area .
as the figure shows , the smaller @xmath21 , the longer it takes to reach the attractor solution .
with wmap 3yr constraint , the attractor solution will be reached only in @xmath80 yr from now .
our upper limits on @xmath21 put pressure on this type of interacting models that , combined with the more generic results of @xcite , show the difficulties that interacting models encounter into explaining the coincidence problem .
if the new wmap 5yr data release sets an even stronger constraint on @xmath21 , the iqm will loose much of its appeal .
et al . _ ,
j. * 607 * , 665 ( 2004 ) ; p. astier _
et al . _ ,
. astrophys . * 447 * , 31 ( 2006 ) ; w.m .
wood - vasey _ et al .
_ , arxiv : astro - ph/0701041 .
d.n . spergel _
et al . _ ,
j. suppl * 170 * , 377 ( 2007 ) ; g. hinshaw _
_ , astrophys
. j. suppl * 170 * , 288 ( 2007 ) .
m. colless _
et al . _ , mon . not .
. soc . * 328 * , 1039 ( 2001 ) ; m. tegmark _
et al . _ ,
d * 69 * , 103501 ( 2004 ) ; s. cole _ et al .
_ , mon . not
* 362 * , 505 ( 2005 ) ; v. springel , c.s .
frenk , and s.m.d .
white , nature ( london ) * 440 * , 1137 ( 2006 ) .
boughn , and r.g .
crittenden , nature * 427 * , 45 ( 2004 ) ; s.p .
boughn , and r.g .
crittenden , mon . not .
. soc . * 360 * , 1013 , ( 2005 ) ; p. fosalba , e. gaztaaga , and f.j .
castander , astrophys .
j. * 597 * , l89 ( 2003 ) ; p. fosalba , and e. gaztaaga , mon . not .
. soc . * 350 * , l37 ( 2004 ) ; j. d. mcewen , p. vielva , m. p. hobson , e. martinez - gonzalez , a. n. lasenby , mon . not .
376 * , 1211 ( 2007 ) ; m.r .
nolta , astrophys .
j. , * 608 * , 10 ( 2004 ) .
p. vielva , e. martnez
gonzlez , and m. tucci , mon . not .
* 365 * , 891 ( 2006 ) .
contaldi , h. hoekstra , and a. lewis , phys .
lett . * 90 * , 221303 ( 2003 ) .
t. padmanbahn , phys . rep . * 380 * , 235 ( 2003 ) ; v. sahni , in _ lecture notes in physics _
( springer , heidelberg , 2005 ) vol .
653 , p. 141 ; j.a.s .
lima , braz .
* 34 * , 194 ( 2004 ) ; l. perivolaropoulos , aip conf . proc . * 848 * , 698 ( 2006 ) .
e. elizalde , j. phys .
a. math.theor .
* 40 * , 6647 ( 2007 ) .
copeland , m. sami , and s. tsujikawa , int .
d * 15 * , 1753 ( 2006 ) .
jerme martin , private communication .
l. amendola , m. gasperini , and f. piazza , phys .
d * 74 * , 127302 ( 2006 ) . c. wetterich , nucl .
b * 302 * , 668 ( 1988 ) ; astron .
astrophys . * 301 * , 321 ( 1995 ) .
a. coley and p. bylliard , phys .
d * 61 * , 083503 ( 2000 ) ; g. farrar and p.j.e .
pebles , astrophys .
j. * 604 * , 1 ( 2004 ) ; y. myung , phys .
b * 626 * , 1 ( 2005 ) ; r - g . cai and a. wang , j. cosmology astropart .
03(2005)002 ; s. del campo , r. herrera and d. pavn , phys .
d * 71 * , 123529 ( 2005 ) ; b. wang et al , phys .
b * 624 * , 141 ( 2005 ) ; d. pavn and w. zimdahl , phys .
b * 628 * , 206 ( 2005 ) ; b. wang et al , phys .
b * 637 * , 357 ( 2006 ) ; h. kim _ et al .
b * 632 * , 605 ( 2006 ) ; b. hu and y. ling , phys .
d * 73 * , 123510 ( 2006 ) ; s. lee , g .- c .
liu , and k .- w .
ng , phys .
d * 73 * , 083516 ( 2006 ) ; b. guberina _
_ , j. cosmology astropart .
phys . 01 ( 2007 ) 012
; r. horvath , phys .
b * 648 * , 374 ( 2007 ) ; m.s .
berger and h. shojaei , phys .
d * 74 * , 043530 ( 2007 ) ; r. mainini , and s. bonometto , j. cosmology astropart .
06(2007)020 ; w. zimdahl and d. pavn , class .
quantum grav .
* 24 * , 5461 ( 2007 ) ; o. bertolami , f. gil pedro and m. delliou , phys . lett .
b ( in the press ) , arxiv : astro - ph/0703462 ; e. abdalla , l.r .
abramo , l. sodre , and b. wang , arxiv:0710.1198 [ astro - ph ] .
f. piazza and s. tsujikawa , j. cosmol .
. phys . * 07 * 004 ( 2004 ) ; b. gumjudpai , t. naskar , m. sami and s. tsujikawa , j. cosmol .
. phys . * 06 * , 007 ( 2005 ) .
l. amendola , m. quartin , s. tsujikawa , and i. waga , phys .
d * 74 * , 023525 ( 2006 ) .
chimento , a.s .
jakubi , d. pavn , and w. zimdahl , phys .
d * 67 * , 083513 ( 2003 ) .
w. zimdahl , d. pavn , and l.p .
chimento , phys .
b * 521 * , 133 ( 2001 ) .
guo , n. ohta , and s. tsujikawa , phys .
d * 76 * , 023508 ( 2007 ) .
g. olivares , f. atrio - barandela , and d. pavn , phys .
d * 71 * , 063523 ( 2005 ) .
brookfield , c. van de bruck , d.f .
mota , and d. tocchini - valentini , phys .
* 96 * , 061301 ( 2006 ) .
peebles and b. ratra , rev .
phys . * 75 * , 559 ( 2003 ) .
k. hagiwara _
d * 66 * , 010001 ( 2002 ) .
copeland , a.r .
liddle , and d. wands , phys .
d * 57 * , 4686 ( 1998 ) .
d. tocchini - valentini and l. amendola , phys .
d * 65 * , 063508 ( 2002 ) ; a.w .
brookfield , c. van de bruck , d.f .
mota , and d. tocchini - valentini , phys .
d * 73 * , 083515 ( 2006 ) .
g. olivares , f. atrio - barandela , and d. pavn , phys .
d * 74 * , 043521 ( 2006 ) .
et al . _ , month . not .
. soc . * 362 * , 505 ( 2005 ) .
m. tegmark _
et al . _ ,
d * 74 * , 123507 ( 2006 ) .
w.j . percival _
et al . _ ,
astrophys . j. * 657 * , 645 ( 2007 ) .
mzaldarr / cmbfast / cmbfast.html .
et al . _ , astrophys . j. suppl .
* 148 * , 195 ( 2003 ) .
a. gelman , and d. rubin , statistical science * 7 * , 457 ( 1992 ) .
a. lewis , s. bridle , phys .
d * 66 * , 103511 ( 2002 ) .
g. olivares , f. atrio - barandela , d. pavn ( 2007 ) .
phys rev d ( submited ) .
http://lambda.gsfc.nasa.gov/ w.j .
et al . _ ,
j. , * 657 * , 51 ( 2007 ) ; w.j . percival _
et al . _ , month . not .
381 * , 1053 ( 2007 ) .
g. schwarz , annals of statistics * 461 * , 5 ( 1978 ) ; a.r .
liddle , month . not .
351 * , l49 ( 2004 ) .
linder and r. miquel , arxiv : astro - ph/0702542 . | interacting quintessence models have been proposed to explain or , at least , alleviate the coincidence problem of late cosmic acceleration . in this paper
we are concerned with two aspects of these kind of models : ( i ) the dynamical evolution of the model of chimento _ et al . _
[ l.p .
chimento , a.s .
jakubi , d. pavn , and w. zimdahl , phys .
rev .
d * 67 * , 083513 ( 2003 ) .
] , i.e. , whether its cosmological evolution gives rise to a right sequence of radiation , dark matter and dark energy dominated eras , and ( ii ) whether the dark matter dark energy ratio asymptotically evolves towards a non - zero constant . after showing that the model correctly reproduces these eras
, we correlate three data sets that constrain the interaction at three redshift epochs : @xmath0 , @xmath1 , and @xmath2 .
we discuss the model selection and argue that even if the model under consideration fulfills both requirements , it is heavily constrained by observation .
the prospects that the coincidence problem can be explained by the coupling of dark matter to dark energy are not clearly favored by the data . |
in the standard model ( sm ) , @xmath3 is dominated by contributions from the gluonic and electroweak penguin ( ewp ) operators , @xmath4 and @xmath5 . in the @xmath6
chiral limit , the @xmath2 matrix elements of the ewp operators @xmath7 , are determined by two 4-quark vevs , @xmath8 and @xmath9 , which also determine the dimension @xmath1 part of the ope of the flavor @xmath10 v - a correlator difference @xmath11 @xcite , where the superscript @xmath12 denotes the sum of spin @xmath13 and @xmath14 components and , with @xmath15 the standard v or a @xmath10 current , the scalar correlators @xmath16 are defined via @xmath17 since @xmath18^{ope}_{d=6}$ ] is strongly dominated by the contribution involving @xmath9 , which vev also dominates the chiral limit @xmath5 matrix element , the extraction of @xmath18_{d=6}^{ope}$ ] is of considerable phenomenological interest , and a number of dispersive and finite energy sum rule ( fesr ) analyses have attempted it @xcite .
@xmath0 decay data plays a key role in these analyses since the spectral function of @xmath19 , @xmath20 , is directly measurable for @xmath21 in non - strange hadronic @xmath0 decays .
explicitly , in the sm , with @xmath22 a short - distance ew correction , @xmath23 , @xmath24 and @xmath25/ \gamma [ \tau^- \rightarrow \nu_\tau e^- { \bar \nu}_e ] $ ] , one has , for the continuum ( non-@xmath26-pole ) part of @xmath27 @xcite @xmath28 dispersive analyses employ the unsubtracted dispersion relation for @xmath19 and require either assumptions about the saturation of the dispersion integral within the range kinematically accessible in @xmath0 decays , or supplementary constraints on @xmath29 for @xmath30 , such as those provided by the weinberg sum rules @xcite and the dgmly @xmath26 electromagnetic ( em ) self - energy sum rule @xcite ( see , e.g. , ref .
@xcite for details ) .
higher dimension ( @xmath31 ) contributions to @xmath32 must also be considered .
these problems are avoided in the fesr approach , which relies on @xmath33 having no kinematic singularities and hence satisfying the fesr relation @xmath34 for any @xmath35 and any @xmath36 analytic in the region of the contour . for sufficiently large @xmath35 ,
the ope should become reliable on the rhs . choosing polynomial weights @xmath36 with degree @xmath37 strongly suppresses ope contributions with @xmath38 . for sub - asymptotic @xmath35 , ope breakdown , or duality violation ( dv )
is expected .
in fact , even for @xmath39 , sizeable @xmath35-dependent deviations between the lhs and ope versions of the rhs are found for the @xmath40 v and a analogues of eq .
[ fesrreln ] @xcite .
these are strongly suppressed for analogues employing pinched weights ( @xmath36 with a zero at @xmath41 ) @xcite , indicating that at scales @xmath42 dvs are localized to the vicinity of the timelike axis . with this in mind , the analysis of ref .
@xcite ( cgm ) employed doubly pinched weights , checking the @xmath35-dependence of the match between the weighted spectral integrals and optimized ope fit as a test of the self - consistency of the assumed neglect of residual dv contributions .
figure 1 shows the resulting residuals , @xmath43/\delta i^w_{opal}(s_0)$ ] , over an expanded @xmath35 range , for the two weights , @xmath44 and @xmath45 of the `` maximally safe '' cgm analysis based on opal data @xcite .
( we focus here on opal data due to a problem with the aleph covariance matrices @xcite which is the subject of ongoing reanalysis . )
@xmath46 are the lhs and rhss of eq .
[ fesrreln ] and @xmath47 the uncertainty on @xmath48 .
it is obvious that residual dvs , though not evident _ within errors _ above @xmath49 , become non - negligible below this point .
small residual dv contibutions are thus expected in the @xmath50 cgm fit window as well . lacking a model for dvs
, analyses such as cgm were unable to estimate the systematic uncertainty associated with neglecting these contributions .
in refs .
@xcite , a model for dv spectral contributions was developed .
the model builds on earlier work in refs .
@xcite and is motivated by large-@xmath51 and regge - based resonance spacing ideas .
the model leads to anstze @xmath52 , @xmath53 , for the v and a spectral functions , where the dv contributions have the form @xmath54 in refs .
@xcite the impact of dvs on previous v - a analyses was investigated using a _ single _ dv ansatz of the form eq .
[ dvmodelform ] for the v - a difference @xmath29 .
this involves the implicit additional assumption that @xmath55 and @xmath56 , allowing the @xmath57-parameter v - a difference to be re - written in the effective @xmath58-parameter form , eq .
[ dvmodelform ] .
we avoid this additional assumption and fit the v and a dv parameter sets separately , as part of a combined v , a fit which also determines the ope parameters @xmath59 , @xmath60 , and the relevant @xmath1 and @xmath57 v and a channel effective condensates .
we find central dv parameter fit values not in good accord with the expectations @xmath55 , @xmath56 .
our analysis employs @xmath36 up to degree @xmath61 , including @xmath40 , which is optimally sensitive to the dv contributions .
the resulting fits provide excellent matches between the opal spectral integrals and optimized ope+dv fit forms for all @xmath36 employed and all @xmath35 down to a fit window minimum @xmath62 .
though so far aimed at extracting @xmath59 , and not optimized for extracting @xmath63 v - a condensates , the analysis nonetheless provides preliminary results for these quantities . since the fits
provide a prediction for @xmath29 for @xmath64 , and hence also above @xmath65 , we can test our results against the weinberg and dgmly sum rules , which constraints have _ not _ been incorporated in performing the fits .
the first and second weinberg sum rules are written in a form with rhss equal to zero ; for the rhs of the dgmly sum rule we employ the @xmath66 chiral limit value @xmath67_{em } /3\alpha_{em}\
, = \ , -0.0109(15)\ { \rm gev}^4 $ ] @xcite .
the results of these tests are shown in figure 2 , with @xmath35 the point beyond which the fitted form of @xmath29 is employed in the relevant spectral integral .
below this point , experimental data are used .
the dotted and solid black lines in the third panel show the central dgmly sum rule rhs and error .
all three sum rules should be satisfied for all @xmath35 in our fit window .
this is evidently the case , giving us good confidence in the results for the fitted ope parameters as well .
as an illustration of our preliminary results , we quote the effective @xmath1 v - a condensate @xmath68 which results from an update of the cgm opal fit incorporating dvs and fitting @xmath69 , @xmath70 , and the @xmath71 , @xmath72 and @xmath73 ( @xmath74 ) fesrs with the cipt scheme for the truncated @xmath75 ope series and @xmath76 .
@xmath68 is defined by @xmath77^{ope}_{d=6}=c_6^{v - a}/q^6 $ ] .
we find @xmath78 , c.f .
the cgm no - dvs maximally safe analysis opal - based value @xmath79 .
values somewhat larger in magnitude , with larger errors , are obtained from analyses excluding @xmath80 , including those employing fopt for the @xmath75 series .
we emphasize that these results are preliminary , and that a dedicated v - a analysis , aimed at reducing the errors , is in progress .
readers noting the @xmath81 difference between the values quoted above and those obtained from the aleph - based analyses of cgm ( neglecting dvs ) and ref .
@xcite ( including dvs approximately ) should bear in mind that the aleph and opal @xmath82 data differ significantly in the upper part of the spectrum , with the opal data agreeing better with expectations based on cvc and recent preliminary babar and snd @xmath82 electroproduction data @xcite .
this work was supported by the natural sciences and engineering research council of canada , the us doe , micinn ( spain ) under grant fpa 2007 - 60323 , the spanish consolider ingenio 2010 program cpan ( csd2007 - 00042 ) , cicytfeder - fpa2008 - 01430 , sgr2005 - 00916 and the programa de movilidad pr2010 - 0284 .
s. peris , b. phily , e. de rafael , _ phys
. rev . lett . _
* 86 * , 14 ( 2001 ) ; j. bijnens , e. gamiz , j. prades , _ jhep _ * 0110 * , 009 ( 2001 ) ; m. knecht , s. peris , e. de rafael , _ phys .
_ * b508 * , 117 ( 2001 ) ; b.l .
ioffe , k.n .
zyablyuk , _ nucl .
phys . _ * a687 * , 437 ( 2001 ) ; k. n. zyablyuk , _ eur .
phys . j. _ * c38 * , 215 ( 2004 ) ; j. roja , j.i .
latorre , _ jhep _ * 0401 * , 055 ( 2004 ) ; c. a. dominguez , k. schilcher , _ phys .
_ * b581 * , 193 ( 2004 ) ; s. friot , d. greynat , e. de rafael , _ jhep _ * 0410 * , 043 ( 2004 ) ; s. narison , _ phys . lett . _
* b624 * , 223 ( 2005 ) ; s. schael _
et al . _ ( aleph ) , _ phys .
rep . _ * 421 * , 191 ( 2005 ) ; j. bordes _ et al .
_ , _ jhep _ * 0602 * , 037 ( 2006 ) ; p. masjuan , s. peris , _ jhep _ * 0705 * , 040 ( 2007 ) ; a.a .
almasy , k. schilcher , h. spiesberger , _ eur .
phys . j. _ * c55 * , 237 ( 2008 ) . | we discuss a preliminary study of the impact of duality violations on extractions from @xmath0 decay data of the @xmath1 vevs which determine chiral limit standard model @xmath2 matrix elements of the electroweak penguin operators . |
as is well known , a so - called excess current @xmath5 appears at large voltages @xmath1 in josephson junctions ( jj ) with a direct conductance,@xcite that is , the current @xmath5 arises in jjs of the s / n / s or s / c / s types , where n denotes a normal metal ( a wire or a film ) and c a constriction .
this means that the current - voltage ( @xmath0-@xmath1 ) characteristics at large @xmath1 ( @xmath9 , where @xmath10 is the energy gap in the superconductors s ) has the form @xmath11 where @xmath12 is the resistance of the jj in the normal state and the constant @xmath5 is the excess current which can be written in the form @xmath13 here , @xmath14 is a numerical factor equal to @xmath15 in the diffusive limit,@xcite and @xmath16 in ballistic jjs with ideal ( fully transparent ) interfaces.@xcite eq .
( [ 1 ] ) also describes the asymptotic behavior ( @xmath17 ) of the @xmath0-@xmath1 characteristics of s / n / n contacts,@xcite where n is a normal metal reservoir . in the latter case ,
the excess current is twice smaller than in the s / n / s jjs .
the excess current @xmath5 is an essential characteristics of s /
n / n or s / n / s contacts which distinguishes them from the tunnel junctions
s / i / n or s / i / s where this current does not arise . if the s / n or n / n interfaces are not ideal ( the transmission coefficient differs from 1 ) , the coefficient @xmath14 in eq .
( [ 1 ] ) can be either positive or negative . that is , an excess @xmath5 or deficit @xmath18 currents arise in this case .
their values depend on the interface transparencies of both interfaces.@xcite the appearance of the excess current at large @xmath1 as well as the non - zero subgap conductance @xmath19 of the s /
n / n contacts at @xmath20 and @xmath21 is explained@xcite in terms of andreev reflections ( ar).@xcite it has been shown in refs . that the zero bias conductance @xmath22 coincides with the conductance in the normal state and has a non - monotonous dependence on the applied voltage @xmath1 or temperature @xmath23 .
similar behavior of the conductance takes place in the so - called andreev interferometers ( see experimental observations in refs . and theoretical explanations in refs . ) .
the andreev reflection implies that an electron moving in the normal metal towards the superconductor is converted at the s / n interface into a hole with opposite spin which moves back along the same trajectory .
physically , this process means that an electron with momentum @xmath24 and spin @xmath25 moving from the n - metal penetrates the superconductor s and forms there a cooper pair , i.e. , it pulls another electron with opposite momentum @xmath26 and spin @xmath27 .
the absence of this electron in the n - metal is nothing else as the creation of a hole with momentum @xmath26 and spin @xmath27 . in the superconductor /
ferromagnet ( s / f ) contacts , the ar is suppressed since the exchange field @xmath3 acting on spins breaks the symmetry of spin directions . de jong and beenakker@xcite have shown that the conductance @xmath28 in ballistic s / f systems is reduced with increasing @xmath3 and turns to zero at @xmath29 , where @xmath30 is the fermi energy . at high exchange energy , electrons with only one spin direction exist in the ferromagnet f so that the ar at s / f interfaces is not possible .
one can expect a similar behavior of the conductance in s@xmath2/n / n contacts , where a `` magnetic '' superconductor with a spin filter s@xmath2 ( see below ) supplies only fully polarized triplet cooper pairs penetrating the n - metal .
it consists of an s / f bilayer and a spin filter fl which passes electrons with only one spin direction , so that one deals with the s@xmath2 superconductor constructed as a multylayer structure of the type s / f / fl . in this case , the conventional ar at the s@xmath2/n interface is forbidden and , therefore , the subgap conductance at low temperatures as well as the excess current may disappear .
as will be shown in this work , the subgap conductance as well as the excess current @xmath5 remain finite in s@xmath2/n / n contacts .
the magnitude of the current @xmath5 and its sign depend on the value of the exchange field in the ferromagnet f. in the considered case of s@xmath2/n / n contacts , the subgap conductance and the excess current occur due to an unconventional ar in which two electrons with parallel spins in the n - film form a triplet cooper pair with the same direction of the total spin .
therefore , the ar at the s@xmath2/n interface is not accompanied by spin - flip ( the hole in the n - wire has the same spin direction as the incident electron ) .
note that , nowadays , the interest in studies of the excess current is revived in the light of recent measurements on s / sm / s jjs with unconventional semiconductor sm ( topological insulator ) in which the josephson effect can occur due to majorana modes ( see recent experimental papers refs . , and references therein ) . in these junctions ,
the excess current also has been observed . on the other hand ,
properties of high-@xmath31 superconductors including the iron - based pnictides have been also studied with the aid of point - contact spectroscopy in which the differential conductance of n / s point contacts has been measured.@xcite a theory of differential conductance of n / s point contacts composed by a two band superconductor with energy gaps of different signs [ @xmath32 has been presented in ref . . in this paper , we calculate the @xmath0-@xmath1 characteristics of diffusive superconductor / normal metal systems of two types .
in the first type of contacts , s@xmath33/n / n , the `` magnetic '' superconductor s@xmath33 is a singlet superconductor s covered by a thin ferromagnetic layer [ see fig .
[ fig : system1a ] ( a ) ] . in this case , both the singlet and the triplet cooper pairs penetrate into the n - wire . in the second type of contacts , s@xmath2/n / n , the magnetic superconductor s@xmath2 consists again of an s / f bilayer which is separated from the n - wire by a spin filter fl [ see fig .
[ fig : system1a ] ( b ) ] .
the spin filter fl is assumed to pass only electrons with spins oriented along the @xmath34 axis ( @xmath35 ) . using the quasiclassical theory ,
we show that in both types of contacts , s@xmath33/n / n and s@xmath2/n / n , the conductance @xmath36 is affected by the proximity effect and the excess ( deficit ) current @xmath5 ( @xmath18 ) as well as the subgap conductance are finite .
we consider an s@xmath2/n / n contact , in which the `` magnetic '' superconductors are formed by a bcs superconductor s ( s - wave , singlet ) covered by a thin ferromagnetic layer f with an exchange field @xmath37 [ see fig .
[ fig : system1a ] ( a ) ] . due to proximity effect
, the singlet component penetrates from the superconductor into the f film , and also a triplet component arises under the action of the exchange field @xmath37 . as is well known ( see reviews refs . ) , in the case of homogeneous magnetization @xmath38 ( @xmath39 ) in the ferromagnet , the vector of the total spin of triplet cooper pairs @xmath40 lies in the plane perpendicular to @xmath38 .
thus , the s / f bilayer with a sufficiently transparent interface can be considered as a `` magnetic '' superconductor with a built - in effective exchange field @xmath41 which has a nonzero projection onto the @xmath34 axis and an effective energy gap @xmath42 ( to be more exact , the condensate wave functions in the f film are analogous to those in a `` magnetic '' superconductor ) .
the magnitudes of @xmath43 and @xmath42 are determined by certain conditions .
for example , in case of thin f and s layers ( @xmath44 , @xmath45 , where @xmath46 are the thicknesses of the f(s ) layers , @xmath47 and @xmath48 ) and a low f / s interface resistance , one has@xcite @xmath49 in case of a high s / f interface resistance , we obtain ( see the appendix [ app : appendix_a ] ) @xmath50 where @xmath51 is the diffusion coefficient in the f film , @xmath52 , @xmath53 is the conductivity of the f film and @xmath54 is the s / f interface resistance per unit area [ see below eqs .
( [ 4 ] ) and ( [ 4 ] ) ] .
the quantity @xmath55 determined by the interface resistance is the so - called subgap or minigap.@xcite in both cases , the effective exchange field @xmath56 may exceed the effective gap @xmath57 without causing a non - uniform state of the larkin
ovchinnikov fulde
ferrel type@xcite because the thicker s film is only weakly affected by the f film .
for example in the famous experiment,@xcite where a long - range triplet component has been observed in a multilayered s / f/f / f/s josephson junction , the curie temperature in a weak ferromagnet f ( pd@xmath58ni@xmath59 ) was about @xmath60 , that is , much larger than the critical temperature of the superconducting transition in the superconductor s ( nb ) with a transition temperature @xmath61 . in principle
, a similar f/s bilayer can be employed as a prototype of the presented s@xmath33 superconductor .
the f layer in s@xmath2/n / n contacts is separated from the n - wire ( or film ) by a filter that passes electrons only with a certain spin direction , say , parallel or antiparallel to the @xmath34 axis [ see fig . [
fig : system1a ] ( b ) ] . as a filter
, thin layers of strongly polarized magnetic insulator@xcite and dyn or gdn films@xcite can be used . /n
/ n contact the superconductor s@xmath33 consists of a bcs superconductor s and a thin ferromagnetic layer ( denoted by f@xmath62 ) , and is connected to a normal metal reservoir n on the right hand side via a normal metal wire n. ( b ) s@xmath2/n / n contact in addition to the case ( a ) , the s@xmath33 superconductor on the left hand side is covered by a spin filter fl that passes electrons only with a certain spin direction , say , parallel or antiparallel to the @xmath34 axis ( indicated by the thick blue arrow ) .
the superconducting phase on the left hand side is @xmath63 .
( c ) sketch ( not to scale ) of a possible experimental realization of the case ( b ) . ] the convenient method to study the system under consideration is the theory of quasiclassical green s functions.@xcite this technique is generalized for the case of ferromagnet - superconductor structures where a non - trivial dependence of the quasiclassical green s functions @xmath64 on spin indices must be taken into account.@xcite in the considered non - equilibrium case , the green s function @xmath64 is a matrix with diagonal matrix elements ( @xmath65 and @xmath66 ) and non - diagonal element ( @xmath67 ) , where the matrices @xmath68 and @xmath67 are the retarded ( advanced ) and keldysh functions , respectively .
all these functions are @xmath69 matrices in the gorkov - nambu and spin spaces . in the n - wire the matrix @xmath64
obeys an equation which looks similar to the usadel equation@xcite ( see also eq .
( 5 ) in ref . )
@xmath70 = 0 \ , , \label{2}\ ] ] where @xmath71 with the diffusion coefficient @xmath51 .
the matrix @xmath72 is a tensor product of the pauli matrices @xmath73 ( @xmath74 ) and the @xmath75 unit matrix @xmath76 , which operate in the particle - hole and spin space , respectively .
the matrix quasiclassical green s function @xmath64 obeys the normalization condition @xmath77 equation ( [ 2 ] ) is complemented by boundary conditions at the interfaces s@xmath33/n and n / n .
they have the form [ see eq .
( 4.7 ) in ref . ,
refs . , and also the recent work ref . ] @xmath78|_{0 } \ , , \label{4 } \\ 2 \bar{r}_{\text{n } } l \check{g } \partial_{x } \check{g } & = [ \check{g}\ , , \hat{g}_{\text{n}}]|_{l } \ , .
\label{4'}\end{aligned}\ ] ] here , the sub - indices @xmath79 and @xmath80 relate to the n / s@xmath33 and n / n interfaces , respectively , while @xmath81 , where @xmath53 is the conductivity of the n - wire , and @xmath82 denote the s@xmath33/n ( respectively , n / n ) interface resistance per unit area .
the matrix @xmath83 describes the electron transmission with a spin - dependent probability @xmath84 .
if the filters let to pass only electrons with spins parallel to the @xmath34 axis , then @xmath85 so that the probability for an electron with spin up ( down ) to pass into the n - wire is @xmath86 .
we assume that @xmath87 with @xmath88 , and the coefficients @xmath89 and @xmath90 are normalized , i.e. , @xmath91 .
note that coefficients @xmath92 are inverted with respect to the coefficients @xmath93 used in refs . .
consider first eq . ( [ 2 ] ) for the keldysh green s function @xmath67 . in the considered one - dimensional case it has the form @xmath94 = 0 \ , .
\label{5k}\ ] ] the keldysh function @xmath67 can be expressed in terms of the retarded and advanced green s functions @xmath68 , and the matrix distribution function @xmath95 , @xmath96 the distribution function @xmath97 determines the superconducting order parameter @xmath10 , whereas the function @xmath98 describes the dissipative current.@xcite we need to know only the distribution function @xmath98 .
multiplying eq .
( [ 5k ] ) by @xmath99 and taking trace we obtain ( employing the normalization condition eq .
( [ 3 ] ) , in particular , the relations @xmath100 ) @xmath101 \partial_{x } n = j \ , , \label{6f}\]]where @xmath102 are , respectively , the diagonal and off - diagonal elements of @xmath68 matrices in the particle - hole space , and we introduced the notation @xmath103 .
the quantity @xmath104 is independent of @xmath105 . integrating eq .
( [ 6f ] ) we obtain @xmath106 where @xmath107 . using the boundary conditions , eqs .
( [ 4 ] ) and ( [ 4 ] ) , we find @xmath108where @xmath109 - \tanh [ ( \epsilon - ev)/2 t ] \big]}$ ] is the distribution function in the normal metal reservoir ( we set the voltage in the s reservoir equal to zero ) , @xmath110 , @xmath111 , and @xmath112 .
the current @xmath0 is expressed via the `` partial '' current @xmath113 as @xmath114 formula eq .
( [ 8j ] ) generalizes eq .
( 13 ) of ref . for the considered case of a spin - dependent interaction and can be applied to the description of contacts with a condensate consisting of singlet and triplet cooper pairs . in the normal state , above the critical temperature of the superconductor s , one has @xmath115 .
thus , we obtain a standard expression for the current per unit area in an n / n / n contact @xmath116 the denominator is the sum of interface resistances and the resistance of the normal n - wire .
the normalized differential conductance of the contacts under consideration @xmath117 at @xmath21 is @xmath118 where @xmath119 is the normalized voltage .
the normalized current @xmath120 is given by the relation @xmath121 and , at large voltage , can be written in the form @xmath122 where @xmath123 is the normalized current through the contact in the normal state .
the normalized excess ( @xmath124 ) or deficit current ( @xmath125 ) is determined by the expression @xmath126 \
, \mathrm{d } v \ , .
\label{13exc}\ ] ] it is valid at arbitrary temperatures because for the function @xmath127 in eq .
( [ 8j ] ) we have @xmath128 for @xmath129 .
the current @xmath130 can be presented as @xmath131 , where @xmath132 is a subgap current and @xmath133 is the contribution from quasiparticles with energies above the gap ; the normalized current in the normal state is @xmath134 .
we see that the excess current is determined by the retarded ( advanced ) green s functions @xmath68 that obey an usadel - like equation .
this equation can be solved in limiting cases .
we consider a contact with a short n - wire ( @xmath135 ) in which the interface resistances dominate ( @xmath136 ) , i.e. , the interface resistances are much larger than the resistance of the n - wire , @xmath137 . in the case of a short contact , the last term in the denominator of eq .
( [ 11dif ] ) and the second term in eq .
( [ 5k ] ) can be neglected so that the usadel equation for the green s functions @xmath68 acquires the form @xmath138 provided that @xmath139 .
we integrate eq .
( [ 14usadel ] ) once over @xmath105 and obtain @xmath140 from the boundary conditions eqs .
( [ 4 ] ) and ( [ 4 ] ) for the retarded ( advanced ) green s functions , we have @xmath141^{r(a ) } \ , , \label{15j } \\
2 \hat{j}^{r(a)}l & = - \bar{r}_{\text{n}}^{-1 } [ \hat{g}_{\text{n } } \ , , \hat{g}(l)]^{r(a ) } \ , . \label{15j'}\end{aligned}\ ] ] subtracting the first equation from the second we arrive at @xmath142^{r(a ) } = 0 \ , , \label{15usadel}\ ] ] where the matrix @xmath143 is a sum of contributions of the n / n and s@xmath33/n interfaces , @xmath144 and @xmath145}$ ] .
the form of matrices @xmath146 and @xmath147 depends on the type of a superconductor .
[ [ magnetic - superconductors_textm . ] ] `` magnetic '' superconductor s@xmath33
. + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + that is , the superconductor s@xmath148 is represented by an s / f bilayer with a thin ferromagnetic layer f. we assume that the exchange field @xmath37 is aligned parallel to the z axis , @xmath149 . in this case , @xmath150 \ , , \label{eq : lambda_a}\ ] ] with@xcite @xmath151^{-1 } |\epsilon + h| \pm [ \zeta^{r(a)}(\epsilon - h)]^{-1 } |\epsilon - h| } { 2 } \
, , \label{g_s_mag}\\ f_{\text{s } \pm}^{r(a ) } & = \frac{\delta \big [ [ \zeta^{r(a)}(\epsilon + h)]^{-1 } \pm [ \zeta^{r(a)}(\epsilon - h)]^{-1 } \big]}{2 } \ , . \label{f_s_mag}\end{aligned}\ ] ] the terms @xmath152 and @xmath153 in eq .
( [ eq : lambda_a ] ) describe the singlet component and , respectively , the short - range triplet component with the total spin of triplet cooper pairs @xmath40 normal to the @xmath37 vector . note that the energy gap @xmath10 and the exchange field @xmath3 in eqs .
( [ g_s_mag ] ) and ( [ f_s_mag ] ) mean the effective @xmath42 and @xmath56 defined in eqs .
( [ 1a])([1b ] ) . in the following , for brevity
, we drop the subindex `` eff '' .
[ [ triplet - superconductor - s_textt . ] ] `` triplet '' superconductor s@xmath2 .
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + this case can be realized with the help of an s / f bilayer with the @xmath37 vector aligned , for instance , along the @xmath105 axis .
the s / f bilayer is assumed to be separated from the n - wire by a spin filter oriented parallel to the @xmath34 axis .
then , @xmath154 \,.\ ] ] the last term describes fully polarized triplet cooper pairs with the @xmath40 vector oriented along the @xmath34 axis .
[ [ bcs - superconductor . ] ] bcs - superconductor .
+ + + + + + + + + + + + + + + + + + + for completeness , we consider also the case of the bcs superconductor which is obtained from the case of a `` magnetic '' superconductor s@xmath33 setting @xmath155 . here , @xmath156 \,,\ ] ] with @xmath157^{-1 } \
f_{\text{s}}^{r(a ) } & = \delta \big [ \zeta^{r(a ) } \big]^{-1 } \,,\end{aligned}\ ] ] and @xmath158 . in order to make the results more transparent , we assume that the parameter @xmath160 is small and both parameters @xmath161 are large ( @xmath162 ) .
these conditions correspond to experimental systems and mean that the s / n interface resistance is much larger than the resistance of the n / n interface and both interface resistances are larger than the resistance of the short n - wire .
then , the solution for a small correction @xmath163 [ where @xmath164 are the quasiclassical retarded ( advanced ) green s functions in the separated n - wire ] is @xmath165 we see that in the lowest approximation in the parameter @xmath160 only the condensate wave function , off - diagonal in the gorkov - nambu space , is changed due to proximity effect .
the correction @xmath166 is small if the parameter @xmath167 is small or , in the case of the s@xmath2/n / n contact , if the parameter @xmath168 is small .
using the known function @xmath169 and eq .
( [ 11dif ] ) , we can readily calculate the normalized conductance @xmath170 at @xmath21 .
thus , we obtain @xmath171^{-1}}|_{\epsilon = v } \ , , \label{17con}\ ] ] with the functions @xmath172^{-1 } |\epsilon + h| + [ \zeta^{r}(\epsilon - h)]^{-1 } |\epsilon - h| \big\ } } { 2 } \ , , & \text{s$_{\text{m}}$/n / n } \ , , \\ \frac { \re\big\ { [ \zeta^{r}(\epsilon + h)]^{-1 } |\epsilon + h| + [ \zeta^{r}(\epsilon - h)]^{-1 } |\epsilon - h| \big\ } } { 2 } \ , , & \text{\text{s$_{\text{t}}$/n / n } } \ , , \\ \re\big\ { \frac{|\epsilon|}{\zeta^{r}_{+}(\epsilon ) } \big\ } \ , , & \text{s /
n / n } \ , , \end{cases}\end{aligned}\ ] ] and @xmath173 ^ 2 + \big [ \re \big\ { \frac{\delta}{\zeta^{r}(\epsilon + h ) } - \frac{\delta}{\zeta^{r}(\epsilon - h ) } \big\ } \big]^2}{2 } \ , , &
\text{s$_{\text{m}}$/n / n } \ , , \\ \frac{\big [ \re \big\ { \frac{\delta}{\zeta^{r}(\epsilon +
h ) } - \frac{\delta}{\zeta^{r}(\epsilon - h ) } \big\ } \big]^2}{2 } \ , , & \text{\text{s$_{\text{t}}$/n / n } } \ , , \\ \big [ \re\big\ { \frac{\delta}{\zeta^{r}_{-}(\epsilon ) } \big\ } \big]^2 \ , , & \text{s / n / n } \ , , \end{cases } \notag\end{aligned}\ ] ] where @xmath174}}$ ] .
equation ( [ 17con ] ) determines the dependence of the normalized differential conductance on the normalized voltage @xmath175 . ) as a function of normalized voltage for the ( a ) s@xmath33/n / n contact , ( b )
s@xmath2/n / n ( in both cases , the parameters are @xmath176 , @xmath177 for the black solid line and @xmath178 for the red dashed line ) , and ( c ) s / n / n contact , where s is a bcs superconductor ( the parameter is @xmath176 ) . note that the quantities @xmath10 and @xmath3 are not the true energy gap and the magnetic field , respectively , but the in eqs .
( [ 1a])([1b ] ) defined effective values ( see also appendix [ app : appendix_a ] ) . ] the first term in the denominator , @xmath179 determines the resistance of the n / n interface , while the second term is proportional to the resistance of the interface between the n - wire and the corresponding superconductor .
the first term in the square brackets , @xmath180 , determines the conductance of this interface due to quasiparticles with energies above the gap , whereas the second term , @xmath181 , is related to the subgap conductance .
we analyze the differential conductance @xmath170 and the @xmath0-@xmath1 characteristics @xmath182 , @xmath183 for contacts of different types .
equations ( [ 17con])([17i_v ] ) allow one to calculate the conductance and the @xmath0-@xmath1 characteristics of contacts under consideration . in fig .
[ fig : difcond2a ] , we show the dependence of the normalized differential conductance @xmath170 on the normalized voltage @xmath184 for the three types of contacts , i.e. , the s@xmath33/n / n contact [ fig .
[ fig : difcond2a ] ( a ) ] , the s@xmath2/n / n contact [ fig .
[ fig : difcond2a ] ( b ) ] , and the s / n / n contact , where s is a usual bcs superconductor [ fig .
[ fig : difcond2a ] ( c ) ] .
note that the dependence @xmath170 for the case of the bcs superconductor coincides with that for the case of a `` magnetic '' superconductor if one sets @xmath155 .
although the function @xmath170 in fig .
[ fig : difcond2a ] ( c ) looks like the voltage dependence of the differential conductance of an s / i
/ n junction ( where i stands for an insulating thin layer ) , it differs from the latter one because this dependence leads to an excess current @xmath5 .
this current is given by the value of @xmath185 in fig [ fig : exc_current_on_h ] ( a ) at @xmath155 ( blue dashed line ) .
the appearance of the excess current is a direct consequence of the fact that the integral @xmath186 $ ] is not zero as it takes place in tunnel s / i / n junctions .
it is seen from fig .
[ fig : difcond2a ] ( c ) that there is a nonzero subgap conductance in the s / n / n contact .
it is caused by a subgap contribution related to the andreev reflection .
this mechanism is also responsible for a zero - bias peak in the conductance that has been observed in early experiments on s / n / sm contacts ( here , sm is a n - doped semiconductor).@xcite theoretical explanations for the observed subgap conductance is given in refs . . in figs .
[ fig : difcond2a ] ( a ) and [ fig : difcond2a ] ( b ) , we plot the voltage dependence of the normalized conductance of the contacts of s@xmath33/n / n and s@xmath2/n / n types for different values of @xmath3 . in both cases , the subgap conductance is not zero , but it is small in contacts of s@xmath33/n / n type if the exchange field @xmath3 is small compared to @xmath10 . the latter property is due to a negligible contribution to the conductance in the subgap region because this contribution is provided by fully polarized triplet cooper pairs the density of which , @xmath187 , decreases with decreasing @xmath3 since @xmath188 .
note that similar results ( nonzero subgap conductance ) were obtained in ref . , where differential conductance of an f/f / s structure has been studied .
however , the case of fully polarized triplet component has not been considered there .
the subgap conductance in another , although similar , systems has been calculated in ref . on the basis of the scattering matrix approach .
the authors considered a half - metal / ferromagnet / superconductor contact in the ballistic regime assuming that the magnetizations in half - metal and ferromagnet are not collinear .
they assumed also that only a single conducting channel exists in the system so that the quasiclassical theory can not be applied to the system .
to some extent , the results obtained in our paper and in ref . differ .
although the subgap conductance @xmath170 calculated in ref . differs from zero , it turns to zero at @xmath189 whereas @xmath190 obtained by us in the present work is finite . a similar system consisting of a half - metallic ferromagnet and a superconductor has been studied in refs .
the authors assumed that these materials are separated by a spin - active interface .
they also obtained the vanishing zero - bias conductance for @xmath191 . in our case ,
the finite @xmath190 is caused by unconventional andreev reflection of triplet cooper pairs induced in the n - wire due to proximity effect .
this ar make the s@xmath33/n interface partially transparent as it occurs in s / n contacts,@xcite .
the zero - bias conductance @xmath190 as a function of @xmath3 is depicted in fig .
[ fig : difcond3 ] for the s@xmath33/n / n and s@xmath2/n / n contacts .
it is equal to zero at @xmath155 in the s@xmath2/n / n contact , where only triplet cooper pairs are present , and has a maximum at @xmath192 .
as mentioned above , at @xmath155 the amplitude of the triplet component turns to zero , and hence the zero - bias conductance vanishes . for the contacts for the s@xmath2/n / n ( black solid line ) ,
respectively , s / n / n contact ( red dashed line)in both cases , the parameter @xmath193 .
note that the quantities @xmath10 and @xmath3 are not the true energy gap and the magnetic field , respectively , but the in eqs .
( [ 1a])([1b ] ) defined effective values ( see also appendix [ app : appendix_a ] ) . ]
we investigate the @xmath0-@xmath1 characteristics of the contacts of the types s@xmath33/n / n and s@xmath2/n / n . [ [ s_textmnn - contact . ]
] s@xmath33/n / n contact .
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + in the considered case of small but finite @xmath194 , the @xmath0-@xmath1 characteristics shows an excess current .
in particular , for @xmath155 we obtain @xmath195 [ or , with dimension , @xmath196 with @xmath179 ] . the excess current increases with increasing the exchange field @xmath3 [ see fig .
[ fig : exc_current_on_h ] ( a ) ] .
the @xmath0-@xmath1 curve has a simple form for the case @xmath155 ( bcs superconductor ) . for small @xmath194 and @xmath198
, we obtain @xmath199 in this case , there is an excess current in the @xmath0-@xmath1 curve ( see fig .
[ fig : exc5 ] ) .
for the ( a )
s@xmath33/n / n and ( b ) s@xmath2/n / n contacts . noticeably is the nonmonotonic behavior of the @xmath0-@xmath1 curve in the s@xmath33/n / n contact .
the excess current in the s@xmath2/n / n contact turns to deficit current at low @xmath7 ( see text ) .
the parameter @xmath194 has the values @xmath200 ( black solid lines ) , @xmath176 ( blue dashed lines ) , @xmath201 ( red dash - dotted lines ) , and @xmath202 ( green dotted lines ) .
the current is normalized to the value of the ohm s law current at the voltage @xmath203 , i.e. , @xmath204 , where @xmath205 with the resistance of the contact in the normal state @xmath12 .
note that the quantities @xmath10 and @xmath3 are not the true energy gap and the magnetic field , respectively , but the in eqs .
( [ 1a])([1b ] ) defined effective values ( see also appendix [ app : appendix_a ] ) . ] ( black solid line ) and @xmath201 ( blue dashed line ) .
the current is normalized to the value of the ohm s law current at the voltage @xmath203 , i.e. , @xmath204 .
the black dotted line indicates the ohm s law , @xmath205 with the resistance of the contact in the normal state @xmath12 .
note that the quantities @xmath10 and @xmath3 are not the true energy gap and the magnetic field , respectively , but the in eqs .
( [ 1a])([1b ] ) defined effective values ( see also appendix [ app : appendix_a ] ) . ]
[ [ s_texttnn - contact . ] ] s@xmath2/n / n contact .
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + using eq .
( [ 13exc ] ) we find the excess or deficit current for small @xmath194 and @xmath3 , @xmath206 where @xmath207 .
one can see that at @xmath208^{3/4}}$ ] , there is an excess current and at @xmath209 the excess current is converted into a deficit current , cf .
[ fig : exc_current_on_h ] ( b ) . as is seen from fig .
[ fig : exc_current_on_h ] , the magnitude of the excess current @xmath5 in the case of the s@xmath2/n / n junction is comparable with the excess current in an s / n / n junction with the same interface resistances .
this means that it can be measured experimentally on existing experimental junctions .
we studied transport properties of `` magnetic '' superconductor / normal metal point contacts of different types , in which both the singlet and triplet cooper pairs are present .
it is shown that , as it takes place in point s / n / n contacts with bcs superconductor , the subgap conductance @xmath4 and the excess current @xmath5 are not zero even if only fully polarized triplet component exists in the n - wire . in this case , the @xmath4 and @xmath5 are caused by an unconventional andreev reflection without spin flip ; the hole moving back along the trajectory of an incident electron with a spin @xmath40 has the same spin direction as @xmath40 .
a similar ar , equal - spin andreev reflection , has been studied in a recent paper,@xcite where a contact between a ferromagnet and topological superconductor with majorana modes has been considered .
we considered two types of contacts , namely the s@xmath33/n / n contact , where both the singlet and triplet component exist , and the s@xmath2/n / n contact , in which only fully polarized triplet cooper pairs penetrate into the n - wire . in both types of contacts , the subgap conductance and the excess
current are present . in the second type of contacts , in s@xmath2/n /
n , these are caused by an equal - spin ar . with decreasing the magnitude of the exchange field @xmath3 the excess current in the s@xmath2/n /
n contact is transformed into a deficit current @xmath18 .
the systems considered by us can be realized experimentally taking into account a rapid progress in preparing s / f nanostructures of different kinds.@xcite the obtained results can be used for identifying the long - range triplet component and in future applications in spintronics.@xcite we appreciate the financial support from the dfg via the projekt ef 11/8 - 2 ; k. b. e. gratefully acknowledges the financial support of the ministry of education and science of the russian federation in the framework of increase competitiveness program of nust `` misis '' ( nr . k2 - 2014 - 015 ) .
we consider an f / s bilayer and show that , under certain conditions , the matrix green s function @xmath210 coincides with that in a superconductor with a built - in exchange field @xmath3 .
we assume that the thickness of the f layer @xmath211 is small so that the condition , @xmath212 , is fulfilled .
then , the usadel - like equation ( [ 2 ] ) in the f region can be integrated over the thickness and we come to eq .
( [ 15usadel ] ) with @xmath213 + { \hat{\tau}_{2 } } f_{\text{s}}}$ ] , where @xmath214 is the green s function in s , @xmath215 .
the exchange field vector @xmath37 is set along the @xmath34 axis .
the subgap energy @xmath55 is defined in eq .
( [ 1b ] ) and we use the matsubara representation .
the matrix @xmath210 is diagonal in the spin - space with elements @xmath216 .
the retarded green s function @xmath65 can be directly obtained from @xmath210 using the relation @xmath217 .
the solution for @xmath210 can be easily found as in sec .
[ sec:2a ] and has the form @xmath218 where @xmath219 . if the resistance of the f / s interface is large enough ( this corresponds to real experiments ) so that the subgap energy @xmath220 is small in comparison with @xmath10 , the solution eq .
( [ a1 ] ) can be written as @xmath221 + { \hat{\tau}_{2 } } \epsilon_{\text{sg}}}{\sqrt{(\omega \pm ih)^{2 } + \epsilon_{\text{sg}}^{2 } } } \ , .
\label{a2}\ ] ] equation ( [ a2 ] ) shows that the green s function in the f layer has the same form as in a superconductor with the energy gap @xmath55 and built - in exchange field @xmath3 .
this equation is valid if the thickness @xmath211 satisfies the condition @xmath222 as follows from this condition , the exchange field @xmath3 can be much larger than the effective energy gap @xmath55 .
65ifxundefined [ 1 ] ifx#1 ifnum [ 1 ] # 1firstoftwo secondoftwo ifx [ 1 ] # 1firstoftwo secondoftwo `` `` # 1''''@noop [ 0]secondoftwosanitize@url [ 0 ]
+ 12$12 & 12#1212_12%12@startlink[1]@endlink[0]@bib@innerbibempty link:\doibase 10.1103/revmodphys.51.101 [ * * , ( ) ] `` , '' in link:\doibase 10.1002/352760278x.ch1 [ _ _ ] ( , , ) pp . @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) link:\doibase 10.1103/physrevb.25.4515 [ * * , ( ) ] \doibase http://dx.doi.org/10.1016/0038-1098(79)90044-9 [ * * , ( ) ] \doibase http://dx.doi.org/10.1016/0921-4534(93)90005-b [ * * , ( ) ] @noop * * , ( ) link:\doibase 10.1103/physrevlett.73.2488 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.73.1416 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.74.602 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.74.5268 [ * * , ( ) ] http://stacks.iop.org/0953-8984/8/i=4/a=001 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.76.823 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.74.1657 [ * * , ( ) ] @noop ( ) , @noop ( ) , link:\doibase 10.1038/nature07081 [ * * , ( ) ] link:\doibase 10.1126/science.1187399 [ * * , ( ) ] http://stacks.iop.org/0953-8984/22/i=46/a=465701 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.105.167003 [ * * , ( ) ] link:\doibase 10.1103/physrevb.83.060510 [ * * , ( ) ] link:\doibase 10.1103/physrevb.84.214521 [ * * , ( ) ] link:\doibase 10.1103/revmodphys.77.935 [ * * , ( ) ] link:\doibase 10.1103/revmodphys.77.1321 [ * * , ( ) ] link:\doibase 10.1063/1.3541944 [ * * , ( ) ] http://stacks.iop.org/0034-4885/78/i=10/a=104501 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.86.3140 [ * * , ( ) ] link:\doibase 10.1103/physrev.175.537 [ * * , ( ) ] @noop * * , ( ) link:\doibase 10.1103/physrev.135.a550 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.104.137002 [ * * , ( ) ] http://stacks.iop.org/0034-4885/74/i=3/a=036501 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.70.853 [ * * , ( ) ] \doibase http://dx.doi.org/10.1063/1.2172647 [ * * , ( ) , http://dx.doi.org/10.1063/1.2172647 ] \doibase http://dx.doi.org/10.1063/1.2787880 [ * * , ( ) , http://dx.doi.org/10.1063/1.2787880 ] @noop ( ) , link:\doibase 10.1103/revmodphys.58.323 [ * * , ( ) ] `` , '' ( , , ) link:\doibase 10.1006/spmi.1999.0710 [ * * , ( ) ] @noop _ _ , the international series of monographs on physics ( , , ) link:\doibase 10.1103/physrevlett.25.507 [ * * , ( ) ] link:\doibase 10.1103/physrevb.86.060506 [ * * , ( ) ] link:\doibase 10.1103/physrevb.86.214516 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.110.047002 [ * * , ( ) ] http://stacks.iop.org/1367-2630/16/i=7/a=073002 [ * * , ( ) ] http://stacks.iop.org/1367-2630/17/i=8/a=083037 [ * * , ( ) ] @noop ( ) , link:\doibase 10.1103/physrevb.92.180506 [ * * , ( ) ] link:\doibase 10.1103/physrevb.92.214510 [ * * , ( ) ] link:\doibase 10.1007/bf00115264 [ * * , ] http://stacks.iop.org/0038-5670/22/i=5/a=r01 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.67.3026 [ * * , ( ) ] @noop * * , ( ) link:\doibase 10.1103/physrevlett.71.1625 [ * * , ( ) ] link:\doibase 10.1103/physrevb.59.12264 [ * * , ( ) ] link:\doibase 10.1103/physrevb.79.024517 [ * * , ( ) ] link:\doibase 10.1103/physrevb.81.094508 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.105.207001 [ * * , ( ) ] http://stacks.iop.org/1367-2630/18/i=2/a=023031 [ * * , ( ) ] @noop ( ) , link:\doibase 10.1103/physrevlett.108.127002 [ * * , ( ) ] link:\doibase 10.1103/physrevb.84.024517 [ * * , ( ) ] link:\doibase 10.1038/nphys3242 [ * * , ( ) ] | we study the @xmath0-@xmath1 characteristics of s@xmath2/n / n contacts , where s@xmath2 is a bcs superconductor s with a built - in exchange field @xmath3 , n represents a normal metal wire , and n a normal metal reservoir .
the superconductor s@xmath2 is separated from the n - wire by a spin filter which allows the passage of electrons with a certain spin direction so that only fully polarized triplet cooper pairs penetrate into the n - wire .
we show that both the subgap conductance @xmath4 and the excess current @xmath5 , which occur in conventional s / n / n contacts due to andreev reflection ( ar ) , exist also in the considered system . in our case , they are caused by unconventional ar that is not accompanied by spin flip .
the excess current @xmath5 exists only if @xmath3 exceeds a certain magnitude @xmath6 . at @xmath7
the excess current is converted into a deficit current @xmath8 .
the dependencies of the differential conductance and the current @xmath5 are presented as a function of voltage and @xmath3 . |
after this paper has been written , paper @xcite appeared , where ( among some other topics ) the addition law for energy - momentum was modified as compared to the one used in @xcite .
this was done to comply with the physical requirement that a set of particles with even sub - planckian energies can have an energy much exceeding the planck energy .
the modification was achieved by simply replacing the planck energy @xmath123 in our units ) for a system of @xmath115 particles with @xmath124 .
+ here we demonstrate that the modified addition law follows from our scheme , by simply adjusting one of our postulates ( postulate iii , ) .
the modification is as follows : we require that for a set of @xmath115 particles the upper bound on both energy and momentum ( normalized by planck energy @xmath0 ) is to be not @xmath125 , but @xmath115 .
this is equivalent to a postulate that the energy composition law for particles ( _ each having planck energy _ ) is a simple addition . in turn , postulate ( @xmath126)for one particle follows from that as a particular case of @xmath127 .
+ thus , if we use thus modified postulate in eqs.([11a],[11b ] ) , we obtain the following value of constant @xmath37 which now depends on the number of particles @xmath115 : @xmath128 as a result , the energy @xmath129 and momentum @xmath130 of a set of @xmath115 particles follow from eqs .
( [ 11a],[11b ] ) ( we restrict our attention to the positive values of @xmath42 ) : @xmath131 @xmath132 the inverse expressions are @xmath133 @xmath134 @xmath135 and @xmath136 represent the conventional sums of the respective individual quasi - energies @xmath137 ( momenta @xmath138 ) ( as in special relativity ) : @xmath139 the respective casimir is found from eqs.([a3 ] ) and ( [ a4 ] ) if we take the mass @xmath29 to be defined the same way in all the regions , from classical to planck s : @xmath144 the composition laws eqs.([a6 ] ) , ( [ a7 ] ) are reduced to the conventional addition laws not only if all the individual energies @xmath140 ( momenta @xmath141 ) are the same @xcite , but also if at least one of the values @xmath145 ( @xmath146 ) .
these laws take especially simple form for two single particles : @xmath147 99 j.kowalski-glikman and s.nowak , hep - th/0203040 j.lukierski and a.nowicki , hep - th/0203065 j.lukierski and a.nowicki , hep - th/0207022 j.rembielinski and k.smolinski , hep - th/0207031 g.amelino-camelia , gr - qc/0012051 ; int.j.mod.phys.*d*11:35-60,2002 g.amelino-camelia , hep - th/0012238 ; phys.lett.*b*510:255-263,2001 j.magueijo and l.smolin , hep - th/0112090 p.kosinski,j.lukierski,p.maslanka and j.sobczyk , hep - th/9412114 s.majid and h.ruegg,phys.lett . *
b313*,357(1993 ) n.r.bruno,g.amelino-camelia,j.kowalski-glikman,hep-th/0107039 j.lukierski , h.ruegg , w.j.zakrzewski , ann.of phys.,*243 * , 90 ( 1995 ) j.kowalski-glikman , hep - th/0107054 j.magueijo and l.smolin , gr - qc/0207085 m. visser , gr - qc/0205093 | we consider an alternative approach to double special - relativistic theories .
the point of departure is not @xmath0-deformed algebra ( or even group - theoretical considerations ) but rather 3 physical postulates defining particle s velocity , mass , and the upper bound on its energy in terms of the respective classical quantities . for a specific definition of particle s velocity
we obtain magueijo - smolin ( ms ) version of the double special - relativistic theory .
it is shown that this version follows from the @xmath0-poincare algebra by the appropriate choice of on the shell mass , such that it is always less or equal planck s mass .
the @xmath0-deformed hamiltonian is found which invalidates the recent arguments about unphysical predictions of the ms transformation . a recent research ( e.g.@xcite,@xcite,@xcite,@xcite,@xcite,@xcite,@xcite ) on the so - called double special relativity not only reexamined its relation to @xmath0-deformed kinematics , but in one specific example @xcite also subjected to criticism physical predictions of one of these theoretical constructs @xcite .
+ it should be mentioned that as early as in @xmath1 , j.lukierski with collaborators @xcite demonstrated that there exist an infinite set of transformations reducing the @xmath0-deformed casimir in majid - ruegg basis @xcite ( used in all the double - special relativistic theories , e.g.@xcite ) to the diagonal form .
in fact , it is possible to show that any of these transformations correspond to a different choice of what one can consider as a definition of the deformed mass .
this makes it difficult , without any additional assumptions , to choose a unique physical theory corresponding to the respective transformation .
this difficulty is emphasized @xcite by what looks like apparent non - physical predictions of one of these constructions @xcite .
+ here we revisit the latter work @xcite , more specifically its treatment of the energy - momentum domain , this time departing not from the group - theoretical point of view , but rather from certain physically justified restrictions ( postulates ) imposed on the classically - defined physical quantities , namely energy , mass , and velocity .
an analogous approach was used in @xcite for a more narrowly defined goal : a study of possible definitions of @xmath0-deformed velocities and their addition laws .
we begin by introducing the postulates defining \i ) the velocity of a particle , ii)its mass to be the same for any scale ( from classical to planck scale ) and therefore based on the relations provided by the momentum sector of classical relativity , iii)the existence of the upper bound ( the planck energy ) on the values of both energy and momentum .
we also retain the upper bound ( speed of light @xmath2 ) on a particle velocity .
+ in what follows we use units where @xmath3 , planck constant @xmath4 , and boltzmann constant @xmath5=1 .
we denote the planck energy ( momentum ) by @xmath0 which in these units is equal to the inverse of the planck length @xmath6 ( @xmath7 ) .
the classical relation between energy @xmath8 and momentum @xmath9 in these units has the following dimensionless form : @xmath10 where @xmath11 . and
@xmath12 is particle s mass .
+ quite analogously we introduce the dimensionless expressions for the physical energy @xmath13 and momentum @xmath14 ( different from the above energy @xmath15 and momentum @xmath16 ) applicable in the region of planck - scale physics @xmath17 following @xcite we write the general functional relation between the classical energy - momentum @xmath18 ( not physical anymore in the planck - scale phenomena ) and its planck - scale counterpart @xmath19 : @xmath20 where the functions @xmath21 and @xmath22 to be defined .
+ to find these functions we use the above postulates ( i)-(iii ) .
the dimensionless velocity of a particle @xmath23 ( compatible with its classical definition in terms of the energy - momentum ) is defined as follows @xmath24 where @xmath25 .
note that in this definition the velocity @xmath26 looks as the one used in the classical case , except that now this velocity @xmath27 while in the classical case @xmath28 next we use the second postulate ( ii ) which defines particle s mass @xmath29 to be _ the same _ in _ all the regions _ ( from classical to planck scale ) , and independent of the velocity definition ( [ 3a])-([3b ] ) .
@xmath30 finally we require ( postulate iii ) that @xmath31 where the equality signs correspond to @xmath32 + we begin with the velocity definition according to eq .
( [ 3a ] ) . upon substitution of this equation into eq.([2 ] ) we obtain @xmath33 this means that @xmath34 inserting eq.([7 ] ) into the definition of mass eq.([4 ] ) we arrive at the following differential equation : @xmath35 ^ 2=0\ ] ] its solution is : @xmath36 where the integration constant@xmath37 to be determined on the basis of the above postulates .
+ as a result , according to eqs.([2]),([6 ] ) the energy - momentum @xmath38 is : @xmath39 @xmath40 the value of the integration constant @xmath37 and the choice of the respective sign in the obtained solution ( [ 10a]),([10b ] ) are dictated by our postulate ( iii ) , eq .
( [ 5 ] ) .
+ to determine both , we notice that since in the classical limit @xmath41 the positive(negative ) values of @xmath15 should correspond to positive ( negative ) values of @xmath42 respectively .
this means the following : @xmath43 @xmath44 taking the limit @xmath45 of eq.([11a]),(eq . [ 11b ] ) and using our postulate iii ) ( eq.[5 ] ) we get @xmath46 inserting this value of @xmath37 into eqs.([10a]),([11a]),([11b ] ) we obtain the explicit expressions for @xmath47 and @xmath42 @xmath48 @xmath49 these expressions reproduce the results obtained in @xcite with the only difference that here @xmath42 is the antisymmetric function of @xmath15 in contradistinction to @xcite .
+ if we use the classical expressions for @xmath50 ( with the same @xmath26 and @xmath29 in all the regions ) @xmath51 and eqs.([10a]),([10b ] ) then we readily obtain ( restricting our attention to the positive region of @xmath42 ) the respective expressions ( cf.@xcite ) for @xmath52 : @xmath53 from ( [ add ] ) follows that the rest energy @xmath54 less than the mass @xmath29 : @xmath55 here the equality sign corresponds to the classical region @xmath56 + from expressions ( [ 13 ] ) , ( [ 14 ] ) we obtain the inversion formulas : @xmath57 @xmath58 if we use classical casimir and expressions ( [ 15 ] ) and ( [ 16 ] ) then the respective casimirs for the energy - momentum in the planck region are @xmath59 @xmath60 where ( @xmath61 ) corresponds to the positive values of @xmath42 and ( @xmath62 ) corresponds to negative values of @xmath42 . solving eqs.([17a]),([17b ] ) with respect to @xmath42 and choosing the correct signs ( according to the positive and negative values of @xmath42 , remembering that both @xmath63 ) we arrive at the following relation ; @xmath64 curves correspond to the values of masses @xmath65,width=226,height=226 ] @xmath66 where the upper(lower ) sign corresponds to @xmath67 respectively .
it is seen that the regions of the positive and negative values of @xmath68 are the same with accuracy to the sign .
the graph of @xmath68 is shown in fig.1 .
+ based on the relation between @xmath52 and @xmath69 [ eqs . ( [ 11a ] ) , ( [ 11b ] ) , ( [ 13]),([14 ] ) ] and on the expressions for classical lorentz boost of @xmath69 in the @xmath70-direction with a velocity @xmath71 ( in the units of @xmath72 )
, we can calculate in an elementary fashion the respective boost relations ( found by magueijo and smolin in @xcite with the help of group - theoretical analysis ) energy- momentum @xmath52 at the planck scale .
we write them in the dimensionless form : @xmath73 @xmath74 here @xmath75 is the kroenecker delta - function and @xmath76 .
since particle s velocity has been defined as @xmath77 , it is not surprising that eqs.([19a],[19b ] ) yield the velocity addition rule , coinciding with the classical relativistic rule : @xmath78 it was shown in @xcite that within the context of @xmath0-poinciana algebra various possible doubly - special relativity constructions can be viewed as different bases of this algebra . in particular , magueijo - smolin basis @xcite is one of such bases . in a more general scheme of things , j.lukierski and collaborators
@xcite demonstrated that all possible double special relativistic constructions differ by a suitable choice of what one defines as an effective mass .
+ here we show that the magueijo - smolin transformation ( obtained here in an elementary fashion with the help of simple physical postulates)contains an additional physical constraint on the mass . this transformation can be derived pro - forma from @xmath0-deformed algebra , and the result explicitly shows that in this case the particle mass @xmath79 . to demonstrate that we write the relations of the classical basis ( denoted here as @xmath80 ) by rearranging the formulas given in @xcite : @xmath81\equiv ae^{\overline{\pi}_0}[1-e^{-\overline{\pi}_0}cosh(\overline{\mu})]\ ] ] @xmath82 where @xmath37 is an arbitrary constant to be determined .
the respective casimirs are @xmath83 and @xmath84 + we can symmetrize expressions ( [ 20a ] ) , ( [ 20b ] ) by introducing the following quantities : @xmath85\ ] ] @xmath86 combining ( [ 23a ] ) , ( [ 23b ] ) and ( [ 20a ] ) , ( [ 20b ] ) , we obtain : @xmath87 @xmath88 by comparing eqs.([24a ] ) and ( [ 24b ] ) with eqs .
( [ 13 ] ) and ( [ 14 ] ) we immediately see that magueijo - smolin basis follows from @xmath0-deformed algebra if and only if @xmath89 in this case casimir ( [ 22 ] ) reads @xmath90 comparing eq.([26 ] ) with casimir given by eq.([17a ] ) @xcite we arrive at the conclusion that the mass @xmath29 used in the latter is @xmath91 in addition , if we use ( [ add2 ] ) then eq.([27 ] ) imposes the following condition on the value of the rest energy @xmath92 : @xmath93 + at the first glance this condition looks ( predicated on the restriction on particle s mass @xmath94 ) as overly restrictive . on the other hand , considering magueijo - smolin transform as a `` free - standing '' transformations we are not forced to have an upper bound on the rest energy @xmath54 in the planck region where @xmath95 ( that is one half of the upper bound on @xmath42 ) .
still , there is a strong argument in favor of adopting the upper bound on particle s mass .
if we would like to be consistent then it seems quite reasonable to expect that all the quantities in the region of planck scales to be bounded from above by planck energy @xmath0 , or bounded from below by planck length @xmath6 .
+ we have derived magueijo - smolin transformation without resorting to group - theoretical approach by simply defining particle velocity @xmath26 , its mass @xmath29 , and the upper bound on the magnitude of momentum - energy @xmath52 in terms of the respective classical quantities .
in particular , the velocity @xmath26 is defined as @xmath77 ( eq.[3a ] ) . if we substitute into this definition the relations between @xmath96 and the respective quantities @xmath97 , eqs .
( [ 23a ] ) , ( [ 23b ] ) ( used in the conventional treatment based on @xmath0-poincare algebra ) we arrive at the value of the velocity @xmath26 which is exactly the right group velocity @xmath98 ( obeying classical addition law ) introduced in @xcite : @xmath99 interestingly enough , the particle velocity is identical in 2 different bases : magueijo - smolin basis and the basis used in @xcite .
+ to complete our elementary treatment of magueijo - smolin transform , we address its critique expressed in @xcite .
it is argued there that a definition of the particle velocity according to hamilton equations results in a paradoxical situation where @xmath100 particles of different masses moving in an inertial frame with the same velocity will have different velocities when viewed from another inertial frame .
the fallacy of this conclusion is due to the fact that the authors of @xcite used a @xmath101-@xmath102 hamiltonian formalism . + it has been demonstrated ( e.g.,@xcite,@xcite ) that a velocity definition in a non - commutative space is dictated by an appropriate choice of a @xmath102 hamiltonian formalism ( see also @xcite )
to this end let us consider dimensionless relativistic phase space variables @xmath103 normalized by the appropriate planck scales and whose commutation relations are : @xmath104&=&\delta_{ij } , \nonumber \\ \lbrack\pi_0,\xi_0\rbrack & = & -(1-\pi_0 ) , \nonumber \\ \lbrack\pi_0,\xi_i\rbrack & = & 0 , \nonumber \\
\lbrack \xi_0,\xi_i\rbrack & = & \xi_i~ , \nonumber \\
\lbrack \xi_0,\pi_i\rbrack & = & -\pi_i , \nonumber \\ \lbrack\pi_{\alpha},\pi_{\nu}\rbrack & = & 0\end{aligned}\ ] ] the @xmath0-deformed hamilton equations then yield ( cf.@xcite ) : @xmath105 @xmath106 here one particle hamiltonian @xmath107 is taken to be the @xmath0-invariant casimir , eq . :
@xmath108 if we use this expression in the hamilton s equations ( [ 31a ] ) , ( [ 31b ] ) we obtain : @xmath109 from eqs.([32 ] ) immediately follows the expression for the velocity ( [ 3a ] ) which was postulated from the very beginning : @xmath110 + another seemingly unphysical prediction(s ) of magueijo - smolin ( ms ) basis , as was pointed out by j.rembielinski and k.smolinski @xcite , is connected with an apparent difficulty in formulating statistical mechanics based on ms basis .
it is argued that one - particle partition function is divergent when @xmath111 .
however this conclusion is based on an assumption that the temperature in the planck region is the same as in the classical region .
this is not true , since the existence of the upper limit on the energy immediately implies that there exist a relation between the temperature ( dimensionless ) @xmath112 in the planck region and its counterpart @xmath113 in the classical region analogous to the relations between energies in these two regions , eq .
( [ 11a ] ) . as a result , it is not difficult to demonstrate that the partition function ( expressed as an integral ) does not have singularities .
+ additional criticism of @xmath114 transformation is connected to the fact that for a large number @xmath115 of identical particles , each of energy @xmath116 their total internal energy in the thermodynamic limit ( @xmath117 ) does not depend on temperature . but this represents not a deficiency of the basis , but on the contrary , its advantage .
in fact , since the temperature is bounded from above by @xmath118 , in the limit of infinite number of particles , whose total energy tends to the respective upper boundary ( @xmath119 ) the respective temperature @xmath120 tend to its maximum that is to 1 , which explains an apparent absence of the dependence of the internal energy on temperature . in this case
the internal energy is simply equal to the temperature , and both are equal to unity ( in the chosen units ) .
+ in conclusion we would like to say that ms basis following from very simple and consistent physical postulates introduced here represents an attractive model for a description of phenomena which might be associated with planck scale physics .
in fact , the imposition of upper bound on the energy - momentum , and even mass ( if we adopt @xmath0-poinciana roots of the basis ) which are in agreement with a major postulate of planck scale phenomena , is the feature which is @xmath121 in any other bases .
still , there are some problems with this ( and to this matter , with any other @xmath0-deformed ) model(s ) . in particular , the commutation relation @xmath122 is not consistent with the well - known string uncertainty relation . |
laser induced particle acceleration has drawn considerable interest among researchers all over the world since the pioneering work by tajima and dawson @xcite .
the acceleration gradient of conventional linear accelerators is of the order of @xmath0 v / cm , however today s state of the art lasers are capable to produce the acceleration gradient many orders of what can be achieved using conventional linacs . in general laser based accelerators can be divided based on the medium in which the acceleration takes place , which can be either vacuum or a plasma .
the vacuum as a medium for particle acceleration has some inherent advantages over plasma medium .
the problems like instabilities are absent in vacuum , it is easier to inject the pre - accelerated particle beam in vacuum as compared to the plasma , collisions of particles with media causing energy loss and beam spreading is ruled out , etc .
thus we will focus on particle acceleration in vacuum in this article .
the relativistic motion of the charge particle in large amplitude electromagnetic ( em ) fields are studied in detail by many authors .
the motion of the charged particle in transverse em wave and the constant magnetic field along wave propagation is studied by roberts and buchsbaum @xcite , which was further extended analytically and experimentally by jorv and trivelpiece @xcite .
more recently the in - depth hamiltonian analysis of the dynamics of charge particle in a circularly polarized traveling em wave is been studied by bourdier and gond @xcite .
various different schemes have been proposed for the acceleration of charge particle in traveling em wave @xcite , some includes the homogeneous magnetic field however some includes the two counter propagating em waves . in this paper
we will consider an alternative method to accelerate protons in vacuum by circularly polarized electromagnetic waves , where the main new ingredient is a phase - modulation of the em wave emphasis would be on understanding the dynamics of proton motion under the proposed scheme .
the next section will briefly describe the proposed scheme followed by the results and discussion .
a circularly polarized traveling wave propagating along the @xmath1 direction is considered .
the electric and magnetic fields of the wave are given by , @xmath2 \label{ex}\]]@xmath3 \label{ey}\]]and the magnetic fields are expressed as , @xmath4 \label{bx}\]]@xmath5 \label{by}\]]with , @xmath6 is the phase modulation function which is given by @xmath7 \label{phase}\]]where , @xmath8 is the so called phase modulation factor which controls the extent of the modulation .
a constant magnetic field is also applied along the direction of wave propagation given by @xmath9 .
the electric fields denoted by eq . [ ex ] and [ ey ] can be generated by introducing an _ electro - optic phase modulator_. this is an optical device in which a element displaying the electro - optic effect is used to modulate the beam of light .
the modulation can be done in phase , frequency , polarization , and amplitude .
the simplest kind of modulator consists of a crystal , such as lithium niobate whose , refractive index is a function of the applied electric field @xcite .
an appropriate electric field along a crystal can be applied in such a way that its refractive index modulates , which eventually will introduce the phase lag in the laser beam .
the phase modulation will also depend on the length of the crystal and other parameters .
designing an accurate phase modulator for a specific problem may be an engineering concern , but for the purpose of this article we will assume that such a problem can be solved satisfactorily . the schematic diagram for the proposed scheme is shown in fig . [ geo ] .
the laser pulse is initially passed through the phase modulator so that the spatial and temporal dependence of the electric and magnetic fields are modified according to eqs .
[ ex ] , [ ey ] , [ bx ] and [ by ] .
this modified pulse is then injected into the accelerating cavity , protons under the influence of this modified laser pulse undergoes the acceleration along the transverse direction .
an exact analytical treatment of the problem seems to be too involved because of the nature of electric and magnetic field profiles .
it would be a non - trivial task to solve the momentum equations corresponding to the field equations ( [ ex])-([by ] ) . in view of this
we have numerically analyzed the dynamics of the particle under the influence of the given field profiles .
as is wellknown , the motion of the relativistic particle is described by the following equations , @xmath10 \\ \mathbf{v } & = & \frac{\mathbf{p}/m_{0}}{\sqrt{1+|\mathbf{p}|^{2}/(m_{0}c)^{2 } } } \\ \frac{d\mathbf{r}}{dt } & = & \mathbf{v}\end{aligned}\ ] ] where , @xmath11 , @xmath12 , @xmath13 and @xmath14 are relativistic momentum , velocity , coordinate and mass of the particle .
the above equations are solved numerically by a standard boris leapfrog scheme where particle motion is decomposed into motion in the electric field and the motion in the magnetic field @xcite .
the particle orbits is calculated by substituting eq .
( [ ex ] ) -([by ] ) into the equation pf motion , and specifying the initial condition for the injection energy , letting the initial velocity be directed along @xmath1 . in rest of the paper
we have used the dimensionless units for all physical quantities .
for all the results presented here , the amplitude of the circularly polarized wave is considered to be @xmath15 ( unless otherwise stated ) , where @xmath16 @xcite . similarly magnetic field
is denoted by @xmath17 .
here , @xmath18 and @xmath19 being the charge and mass of electron , @xmath20 and @xmath21 is the amplitude of electric and magnetic field in si units with @xmath22 being the laser frequency .
the dimensionless space and time are taken in units of the wave number @xmath23 and the angular frequency @xmath22 respectively .
in this article our main focus is to understand the dynamics of a single proton with energy 1 kev ( unless otherwise stated ) , injected along the propagation direction ( @xmath1 ) of a phase modulated circularly polarized wave with amplitude @xmath15 and a constant magnetic field with magnitude of @xmath24 ( fig .
[ geo ] ) along wave propagation .
although it will be clear later in the article that this scheme is equally valid for the proton beam having some energy spread .
the central theme of the proposed scheme is the introduction of the so called _ phase modulation factor ( @xmath8 ) _ which can be expressed as the ratio of the phase modulation frequency ( @xmath25 ) to wave frequency @xmath22 i.e. @xmath26 .
next we will see how the value of @xmath8 affects the resulting dynamics of the particle .
the time evolution of the transverse kinetic energy of the particle for the phase modulation factor @xmath27 with @xmath28 and @xmath29 are presented in fig .
[ transke ] ( a ) .
it is observed that the proposed scheme of acceleration works well only when the wave frequency is some harmonics of the modulation frequency . a large deviation from this condition destroy the acceleration mechanism completely , fig .
[ transke ] ( b ) . furthermore , as can be seen from fig .
[ transke](a ) , the efficiency of the scheme deteriorate gradually with higher harmonics of the phase modulation frequency . with @xmath30 2,3 , and 4 being an integer , similarly the case when @xmath31 is not an integer is also presented ( b ) .
the amplitude of the em wave @xmath32 = 0.1 and constant magnetic field along wave propagation @xmath33 = 1.0 is considered . ]
it would be interesting to analyze the underlying principle for the success of this acceleration scheme .
it turns out that the magnetic field as well as relativistic effects plays only a minor role in this regard .
thus , for the sake of simplicity let us consider the following equation of motion , @xmath34 where @xmath35 and @xmath36 is the mass and charge of the particle . only the temporal dependence in eq .
[ modeq ] is considered for the purpose of analysis , which is valid assumption since the dynamics is mostly independent of the spatial coordinates . now using the field profile given by eq .
[ ex ] and phase modulation function given by eq .
[ phase ] , this simple equation of motion ( eq .
[ modeq])can be integrated to give @xmath37d\tau ^{\prime } + c \label{veloint}\ ] ] where @xmath38 is time in dimensionless units and @xmath39 is a constant of integration . for simplification
we have chosen @xmath40 and @xmath41 .
the solution of the above integral ( eq . [ veloint ] ) for different values of @xmath8 are shown in fig .
it is quite clear that if @xmath27 , with @xmath31 being an integer the temporal evolution of velocity is very much regular and periodic ( fig .
[ velo ] a , c ) , however for the values when @xmath31 is not an integer the solution of the integral is irregular , ( fig . [ velo ]
the observation is that the success and failure of the proposed acceleration scheme mostly depend on the property of the integral given by eq .
[ veloint ] .
it is observed that acceleration works only when the frequency of the electromagnetic wave is some harmonics of the phase modulation frequency , i.e. @xmath27 with @xmath31 being an integer . with different values of @xmath42 ( a ) 0.25 ,
( b ) 0.29 , ( c ) 0.33 and ( d ) 0.42 is presented . considered . ] . ]
now let us focus on the trajectories of the particle under the influence of the phase modulated em wave ( eq . [ ex ] , [ ey ] , [ bx ] and [ by ] ) and external magnetic field ( @xmath9 ) .
the time evolution of the particle trajectory along 3 space dimensions is presented in fig .
excurs ( a ) .
here we have considered the value of @xmath43 ( the same results holds qualitatively for other valid values of @xmath44 i.e. @xmath45 and @xmath46 ) .
the longitudinal displacement along @xmath1 is mainly governed by the energy at which the particle is being injected in the cavity ( magenta curve ) , however oscillatory motion along @xmath47 direction ( red curve ) is because of the presence of the external magnetic field along @xmath1 direction . as can be infered from the fig .
[ excurs ] ( b ) and ( d ) , the biggest problem with the acceleration scheme is the excursion along the @xmath48 direction which is much larger than the displacement along the @xmath1 direction of propagation . in this scenario
the proposed scheme would be impossible to implement in practice because of the excursion along transverse direction is too large . in order to deal with this problem
a small detuning parameter @xmath49 can be added to phase modulation factor @xmath8 , which would be helpful to confine the particle orbits in the x - y plane .
let us examine the trajectory of the particle after the addition of a small detuning parameter @xmath50 which modifies the phase modulation factor @xmath8 from 0.5000 to 0.5001 .
it can be observed from fig .
detune(a ) that the kinetic energy of particle increased by a factor of about 2 as compared to case when no detuning parameter is present ( fig .
transke ) .
furthermore the trajectory of the particle is modified significantly after the introduction of the detuning parameter .
as can be seen , the particle trajectory in the x - y plane ( fig . [ detune](d ) ) is closed which make it possible for the acceleration scheme to work in practice . .
] it should be noted that the selection of the detuning parameter @xmath49 is very crucial for the success or failure of the acceleration mechanism .
the value of @xmath49 must be small enough to make the scheme work in favor of acceleration , i. e. we must still have the wave frequency to approximately be a harmonic of the phase modulation frequency , such that the foundation of the scheme is not destroyed , see fig .
[ transke ] ( b ) .
moreover , @xmath51 can not be too small , as it must be large enough to prevent the large excursions ( fig .
[ excurs ] b , d ) .
the effect of the detuning parameter on the energetics of the particle is presented in fig .
[ ener ] . as can be seen from this figure , reducing the detuning parameter from @xmath52 to @xmath53 increases the efficiency of the acceleration significantly . on the other hand ,
the transverse kinetic energy of the particle is directly related to the transverse excursion of the particle . for lower values of the @xmath49
the particle orbits are larger such that the electric field of the wave tend to do more work for each orbital motion of the particle .
as the detuning increases the particle orbit becomes shorter and shorter , resulting in lower gain in energy from the wave .
apparently there is an optimum detuning at which one is able to gain maximum energy per orbital cycle of the particle . .
the resulting phase modulation factor would be @xmath54 with @xmath55 . ]
direction . ]
so far we have presented all the results with constant magnetic field of amplitude @xmath56 .
it can be understood that the magnetic field is also responsible along with the detuning parameter to restrict the excursion of particle in transverse direction . in view of this
it would be interesting to see how the acceleration efficiency varies with the applied constant magnetic field . the temporal evolution of the energy varying the constant magnetic field is presented in fig .
the detuned phase modulation parameter in this case is chosen to be 0.5001 .
it is observed that the proton can now be accelerated to energies of about 386 mev when the applied magnetic field strength is 0.30 .
furthermore , there seems to be an optimum magnetic field for the maximum acceleration of the particle . this behavior can be explained on the basis of the larmor radius of the particle which varies with the applied magnetic field .
the magnetic field should be strong enough to bend the particle to avoid large excursion and should not be so large such that the particle orbit is very small and the resulting energy gain in one orbital motion is small .
the particle orbit for the magnetic field @xmath57 is shown in fig .
[ orbit ] . as can be seen from fig .
[ orbit ] the excursion in the transverse direction is controlled .
thus in this case , particle acceleration using a phase modulated electromagnetic wave propagating along a constant magnetic field seems to be possible in a real scenario .
order polynomial fit which is used in calculating the energy distribution function of output beam.,title="fig:",width=168,height=288 ] order polynomial fit which is used in calculating the energy distribution function of output beam.,title="fig:",width=168,height=288 ]
here we have focussed on the understanding of the underlying dynamics of the acceleration .
we have established that the phase modulated circularly polarized wave can accelerate particles in the presence of a constant magnetic field .
there are optimum conditions on the detuning parameter and the magnitude of the magnetic field , in order to gain maximum energies .
however , for the purpose of applications , the spectral properties plays a crucial role , in addition to the maximum energy . in view of this
it would be important to understand how this scheme works for a particle beam with an energy spread , instead of just a single particle . in order to shed some light on the acceleration dynamics of the particle beam
we have separately simulated the single particle motion when injected with different energies , which can be perceived as the energy spectral spread of the beam .
the time evolution of the transverse kinetic energy of single particle is presented in fig .
[ inj](a ) for different injection energies .
the phase modulation factor is considered to be @xmath58 and the magnitude 0.5 is considered for the constant magnetic field . it can be observed that even if the injection energies are varied an order of magnitude , the output energy of the beam is not very much affected , keeping the scheme functional . the most important property of the particle beam is its energy distribution function .
the energy distribution function of the output beam is calculated in fig .
[ inj](b ) by considering the input beam having gaussian energy spectrum with peak energy of 600 kev and fwhm of 166 kev , the energy spread ( @xmath59/@xmath60 ) of the input beam is about 28% .
the peak energy of the output beam is observed to be 167 mev with fwhm of 5 mev , with the energy spread about 3% .
the inset in fig .
[ inj](b ) shows the variation of output energy with the injected particle energy .
circles denotes the actual numerical value which is fitted with a 10@xmath61 order polynomial ( solid line ) in order to find the energy spectrum .
the decline in the output particle energy with input energy can be explained on the basis of the interaction time of the particle with the em wave .
the faster the particle , the lesser is the interaction time with the em wave , and hence the energy transfer to the particle . as we have observed in fig .
[ inj](b ) , the output particle energy is more or less independent of the injected particle energy , which is apparently visible in the almost monoenergetic energy distribution of the output beam .
a further evaluation of the feasibility of the proposed acceleration scheme need to go beyond the 1d - variations of the electromagnetic fields . in particular , it is clear that picking parameter values corresponding to intense lasers give succesful results for the maximum particle energies .
however , a potential limitation is that large transverse particle excursions exclude the use of very focused pulses , in which case the available laser intensity drops accordingly .
to some extent this may be remedied with a strong value of the static magnetic field , which limits the transverse excursion . in practice ,
magnetic field stengths well beyond @xmath62 @xmath63 is needed if the system works in the optical laser regime , which makes this regime less attractive .
decreasing the wave frequency in the scheme reduces the need for extreme magnetic field strengths , since we may allow for somewhat larger particle excursions .
the optimal frequency regime may lie in the infra - red regime or lower , but a full 3d analysis is needed to optimize the parameters in a realistic scenario .
this remains a project for further research .
this work is supported by the baltic foundation , the swedish research council contract # 2007 - 4422 and the european research council contract # 204059-qpqv .
this work is performed under the _ light in science and technology _ strong research environment , ume university .
it should be noted that the @xmath32 is normalized by electron mass , though the dynamics of the proton will be governed by the field amplitude which is about 1836 times less than @xmath32 , resulting in non - relativistic energy regime . | the acceleration of charged particles , producing collimated mono - energetic beams , over short distances holds the promise to offer new tools in medicine and diagnostics . here , we consider a possible mechanism for accelerating protons to high energies by using a phase - modulated circularly polarized electromagnetic wave propagating along a constant magnetic field .
it is observed that a plane wave with dimensionless amplitude of 0.1 is capable to accelerate a 1 kev proton to 386 mev under optimum conditions .
finally we discuss possible limitations of the acceleration scheme . |
we consider numerical inversion of a forward deformation vector field ( dvf ) from one image to another . very often the inverse dvf is needed along with the forward dvf to map medical images , structures , or doses back and forth throughout the process of 4d image reconstruction and adaptive radiotherapy @xcite .
the inverse dvf may be obtained in different ways , such as through deformable registration with swapped inputs , simultaneous registration in both directions , or inverting the forward dvf from the reference image to the deformed target image . the latter option ( inverting the forward dvf ) is often preferred in clinical applications , due to several reasons :
inversion is typically faster , empirically ; image quality can be quite different for the reference and target image sets , which may make the other approaches more error - prone @xcite ; and inversion can ensure consistency between forward and inverse dvfs @xcite .
previously , chen _ et al . _
@xcite developed a fixed - point iteration method for dvf inversion . in this study
, we aim to advance the dvf inversion approach further by improving its convergence behavior , in terms of convergence region and rate , using a feedback control .
the problem of dvf inversion can be framed as follows .
the reference and target images , denoted by @xmath3 and @xmath4 , respectively , can be related to one another by two non - linear transformations . the forward transformation , @xmath5 , maps the voxels of the reference image , @xmath3 , onto those of the target image , @xmath4 , via the forward deformation vector field , @xmath6 : @xmath7 where @xmath8 is the 3d displacement of the reference voxel at @xmath9 , and @xmath10 is the image domain .
conversely , the backward transformation , @xmath11 , maps the voxels of @xmath4 back to @xmath3 , @xmath12 via @xmath13 , the reverse dvf .
the problem of dvf inversion is to obtain @xmath14 given @xmath15 .
the two transformations are the inverse of each other , i.e. , @xmath16 consequently , the forward and backward dvfs satisfy the _ simultaneous inverse consistency condition _ : [ eq : consistency - condition ] @xmath17 where @xmath18 .
inverse consistency is of great importance to deformable registration and estimation of 4d dose accumulation , among other biomedical applications @xcite .
the inverse consistency condition is commonly incorporated in deformable registration processes .
for instance , christensen and johnson @xcite formulate image registration using objective functions symmetrically between the two images in both matching and regularization terms .
et al . _
@xcite also treat the two images symmetrically , and use inverse consistency in approximating the unknown inverse fields .
additional related studies on employing the consistency condition in simultaneous estimates of the forward and inverse dvf can be found in the survey by sotiras _
et al . _
@xcite .
the study reported in this paper follows and improves upon the work of chen _ et al . _
the precursor work presented a fixed - point iteration method for dvf inversion , with regard to inverse consistency condition .
the significance of that work lies not only in the simple iterative process , but also in the corresponding convergence condition .
assuming @xmath6 is given , chen s iteration proceeds as , @xmath19 the initial guess , @xmath20 , is set to zero ; i.e. , @xmath21 . negating the forward dvf used to a prevailing approach for inverse dvf computation , but the resulting inverse estimate , @xmath22 , does not in general satisfy inverse consistency .
this common misconception was made clear and amended by the fixed - point iteration solution of ( [ eq : fpim ] ) . with fixed - point method ,
the convergence behavior of the iterative inversion process can be analyzed , which is a substantial advancement from previous methods which solely relied on empirical studies .
a sufficient convergence condition for ( [ eq : fpim ] ) is the contraction condition on @xmath6 : @xmath23 where @xmath24 is a well - defined distance metric in the 3d image domain , and @xmath25 is a lipschitz constant , @xmath26 .
the convergence behavior of chen s iteration depends passively on this condition , which is not always met in clinical cases with large deformation . in this study
, we introduce an iterative method with an active feedback control mechanism . at each step of the iteration
, we compute a residual which measures the inconsistency between the forward dvf and the iterative inverse estimate , see ( [ eq : residual - r ] ) .
the residual is incorporated into the next iterate after being modulated by the feedback control .
the feedback control provides an extra handle for controlling and improving the convergence behavior .
the rest of the document is organized as follows . in
, we describe the new iterative method with feedback control , introduce a simple feedback control mechanism , and provide the underlying principle . in , we make experimental assessment of the new method with an analytic dvf pair and with numerical dvfs obtained via the 4d extended cardiac - torso ( xcat ) digital anthropomorphic phantom @xcite . in , we conclude the presented work and give additional remarks on extended feedback control .
an iterative method with feedback control is first introduced for numerical dvf inversion .
an analysis is then provided for steering the feedback mechanism to improve convergence behavior . at each iteration step , we get an iterative estimate , @xmath27 , of the inverse dvf , @xmath14 . we use the residual with respect to the consistency condition of ( [ eq : consistency - condition ] ) as the feedback : @xmath28 the residual can be obtained at each iteration step .
this computationally available quantity allows us to monitor and control the ( unknown ) estimate error , @xmath29 which is to be reduced to zero , or sufficiently close to zero , via the iteration process .
the residual @xmath30 is small when the error @xmath31 is small , and @xmath31 is zero when @xmath27 is equal to the inverse satisfying the consistency condition ( [ eq : consistency - condition ] ) .
feedback control is introduced in the iteration process as follows . the residual for the @xmath32-th iterate , @xmath33 ,
is calculated , modulated by a nonsingular matrix @xmath34 , and used as an incremental correction , @xmath35 where @xmath36 is a @xmath37 matrix , local to each @xmath38 .
feedback control is spatially homogeneous when @xmath34 does not vary with @xmath9 .
there are many formulations or reformulations of a fixed - point equation .
the formulation in ( [ eq : fpim - general ] ) is a general formulation for fixed - point iterations , with feedback control @xmath34 modifying and transforming the residual before it is fed back to @xmath39 . in this study , we focus on the feedback mechanism in its simplest form with a single parameter .
we let @xmath40 , where @xmath41 is the identity matrix , and @xmath42 is a scalar which may be referred to as the relaxation parameter .
the iteration process thus takes the simple form : @xmath43 when @xmath44 , the iteration of ( [ eq : fpim - single - parameter ] ) reduces to chen s iteration of ( [ eq : fpim ] ) .
hereafter , we refer to the iteration procedure by chen _
_ as the iteration with feedback control @xmath45 .
we first demonstrate how the feedback control can be designed for inverting the analytic dvfs introduced by chen _
et al . _
the forward and backward dvfs are expressed by the following closed - form formulas over a 2d spatial domain : [ eq:2d - analytic - dvfs ] @xmath46 where @xmath9 denotes cartesian coordinates in the 2d spatial domain @xmath47^{2}$ ] , and @xmath48 is the corresponding angular coordinate in the polar representation , @xmath49 .
the dvf functions have two constant parameters , @xmath50 and @xmath51 .
illustrates a source image and two target images ; the latter are related to the former by the analytic dvfs of ( [ eq:2d - analytic - dvfs ] ) .
it is straightforward to verify that the analytic dvfs satisfy the consistency condition of ( [ eq : consistency - condition ] ) . when @xmath52 , there is no deformation .
when @xmath53 and @xmath54 , the displacement ( forward or backward ) at @xmath9 is proportional to @xmath9 . with @xmath55 and @xmath56 ,
the deformation at @xmath9 remains in the radial direction , but its magnitude varies with @xmath50 and @xmath57 ; when @xmath50 is close to or larger than @xmath58 , the deformation becomes large at and near angles where @xmath59 .
the iteration with control @xmath45 converges slowly when @xmath50 approaches @xmath58 from below , and becomes divergent when @xmath50 is greater than @xmath58 .
in fact , the contraction condition ( [ eq : contraction - condition ] ) no longer holds at all @xmath48 values when @xmath60 . 1 iteration 5 iterations 10 iterations + to improve convergence , we begin by taking a closer look at how errors propagate through the iteration process .
errors in two successive iteration steps are related by the following propagation equation : @xmath61 \cdot { \mathbf{e}}_{k } ( { \mathbf{x}}).\ ] ] our objective is to suppress the iterative error ( [ eq : error - propagation - customized ] ) at every step and make it vanish at a faster pace .
when @xmath45 and @xmath60 , the error propagation factor is greater than one at and near angular locations where @xmath62 .
wherever the factor is greater than @xmath0 , the iterative errors are magnified instead of suppressed at each iteration .
this explains the divergent behavior with @xmath45 at certain polar angles when @xmath63 , as illustrated in . when @xmath45 and @xmath64 , the propagation factor is less than @xmath0 at any @xmath48 , and the iteration converges in theory .
but as @xmath50 approaches to @xmath58 from below , the iteration becomes slower .
further analysis , provided in , shows that the single - parameter feedback control can be employed to achieve convergence over the entire image domain with this analytic dvf pair .
we claim specifically that , for any @xmath65 , there exists some @xmath66 which satisfies the following necessary and sufficient convergence condition : @xmath67 under this condition , the iteration ( [ eq : fpim - single - parameter ] ) converges over the entire image domain for any @xmath65 , with arbitrary initial guess . for @xmath68 , iteration with the feedback control
@xmath45 fails to meet the necessary and sufficient condition . among the feasible values for the parameter @xmath42
, one may find , by the propagation factor in ( [ eq : error - propagation - customized ] ) , that some values of @xmath42 render faster convergence than others .
in particular , we find the optimal value , @xmath69 we provide the proofs of ( [ eq : error - propagation - customized ] ) , ( [ eq : specific - mu - condition ] ) and ( [ eqn : optimal - mu - specific ] ) in . in this section
, we will describe the basic principle , objectives and approach for feedback control , taking into consideration that , in practice , the forward dvf is provided in numerical values over a discrete spatial domain , not in a closed - form expression . the design of feedback control is based in principle on the error propagation analysis . for the iteration of ( [ eq : fpim - single - parameter ] ) with a single control parameter @xmath42 , the error propagation equation may be described in the first order form , @xmath70 where @xmath41 is the identity matrix , @xmath71 is a location in @xmath72 $ ] , @xmath73 denotes the jacobian of @xmath15 evaluated at @xmath74 , and @xmath75 denotes the error propagation matrix . in order to express the error propagation matrix in a first - order form , we assume that @xmath15 is smooth almost everywhere over the spatial domain and the jacobian can be well approximated from the numerical values of @xmath15 . the first control objective is to make the propagation matrix @xmath76 a contractor over the image domain of interest , which implies that the convergence region must span the entire image domain . when the feedback control , @xmath36 , is limited by a specific mechanism structure , one may consider making the convergence region in the image domain as large as possible . the second objective is to make the convergence as fast as possible .
we describe in particular the control design approach with the single - parameter control mechanism .
first , we find the feasible values of @xmath42 over which @xmath76 is a contractor over the spatial domain of interest , @xmath77 where @xmath78 is the spectral radius of the propagation matrix .
if the feasible set is not empty , we next locate the values of @xmath42 that reach the fastest convergence speed possible . in
, we have demonstrated how to achieve these objectives with analytic dvfs . in
, we will introduce a simple approach to locate @xmath42 in practical situation where the dvf is numerically provided , without a closed - form expression .
we present experimental results for assessing the new iterative method .
first , we will describe the data sets and assessment measures .
two particular sets , referred to as set a and set b for convenience , were used for evaluating the single - parameter feedback control process of ( [ eq : fpim - single - parameter ] ) .
data set a is analytically designed by chen _
et al . _
it includes a reference image and a deformed ( target ) image , in 2d , as shown in .
the reference image is binary - valued ; the intensity is one over the concentric rings , and @xmath79 otherwise .
the dvf from the reference to the target image follows the analytic expression ( [ eq:2d - analytic - dvfs ] ) . in the experiments ,
the dvf parameter @xmath51 is set to @xmath80 , and the parameter @xmath50 takes values below or above @xmath58 ( which is the critical point for chen s iteration from converge to diverge ) .
data set b consists of two images of a digital lung phantom generated using xcat @xcite .
the reference image is at the end - of - expiration phase and the target image is at the end - of - inspiration phase .
the forward dvf , from the reference to the target image , was provided by xcat .
respiratory motion was generated by two parametrized sinusoidal curves : one for diaphragm motion in the si direction , and one for chest surface motion in the ap direction .
the curve and surface parameters were set to the xcat default values , with a respiratory cycle , and asymmetric sinusoidals with respect to inspiration and expiration duration .
peak - to - peak displacements at the diaphragm and chest surface were set to and in si and ap directions , respectively .
the large displacements were chosen in order to test the convergence behavior of the iteration methods in a challenging condition that may be expected in a clinical scenario @xcite .
voxel resolution was set to @xmath81 mm in all three dimensions .
the xcat phantom deformation fields are mostly smooth over the image domain , except at certain surface boundaries , such as around the spine .
we use two kinds of schemes for assessing the effect of feedback control on convergence behavior .
the first kind of scheme shows the spatial variation of errors using images or error maps .
the second kind scheme uses scalar quantities to summarize estimate errors in the worst case or in aggregation . for dataset
a , visual assessment of spatial variation of errors is materialized in two ways . in one ,
the proxy target and references images are constructed by the inverse dvf estimate under evaluation . in the other ,
a map of pixel - wise residuals is overlaid on the target image .
the scalar assessment scheme comprises the maximal residual error and the maximal absolute error at each iteration , [ eq : max - errors ] @xmath82 where @xmath10 is the image domain , @xmath83 ^ 2 $ ] . for dataset b , the scheme for assessing the spatial variation of errors is similar to that for dataset
a , except that proxy images are constructed for only the target phase , for the following reason . unlike the ideal case with dataset a
, local transforms of image voxels by the dvfs may not be one - to - one numerically , which happens to be the case with the backward transform for dataset b. this numerically ill condition makes the constructed reference images via numerical interpolation much more susceptible to numerical artifacts , especially around the outer body interface .
the scalar assessment scheme is limited to the measure of residual errors , as the absolute errors are not obtainable with the exact inverse unknown .
taking into account the presence of discontinuities at certain boundaries and other sources of variation , residual errors are assessed via percentiles at different levels , defined as , @xmath84 where @xmath85 is the normalized cumulative histogram of @xmath86 , and the image domain @xmath10 excludes voxels outside the body . in particular
, we use percentiles with @xmath87 and @xmath88 .
we present experimental results , in , with the analytical dvf of ( [ eq:2d - analytic - dvfs ] ) , at two dvf parameter settings .
the parameter @xmath51 is set to @xmath80 in both settings and @xmath50 is set to @xmath89 in one setting and @xmath90 in the other .
1 iteration 5 iterations 10 iterations + @xmath91 + @xmath92 1 iteration 5 iterations 10 iterations + @xmath91 + @xmath93 1 iteration 5 iterations 10 iterations + @xmath91 + @xmath92 1 iteration 5 iterations 10 iterations + @xmath91 + @xmath93 when @xmath94 , the iteration of ( [ eq : fpim - single - parameter ] ) converges for any @xmath66 .
show the iterative estimates of the reference and target images respectively at intermediate steps of the iteration process .
the iteration of ( [ eq : fpim - single - parameter ] ) with feedback control @xmath95 recovers the reference and target images in at most @xmath96 steps , whereas the iteration with feedback control @xmath45 takes more steps .
a quantitative confirmation of the speedup , shown in [ fig : result - analytical_phantom - b_pt_4 ] , shows the comparison in convergence rate among the feasible values .
the iteration with @xmath97 is at least two times faster than with @xmath45 .
show the results when the deformation is larger .
the iteration with @xmath45 diverges , whereas iteration with @xmath98 recovers the reference and target images successfully in at most @xmath99 steps .
shows the critical change from divergence to convergence .
provide numerical evidence that errors decrease faster when the feedback control parameter @xmath42 is set to the analytically optimal value in ( [ eqn : optimal - mu - specific ] ) .
we present experimental results for the xcat phantom .
shows residual errors overlaid on ct slices .
the residual errors are much smaller with the feedback control @xmath100 , especially around the diaphragm . shows the spatial distribution of errors by means of the constructed target image .
the target image is constructed by deforming the reference image with inverse estimate .
the constructed target image with @xmath100 exhibits substantially smaller errors compared to that with @xmath45 . shows the residuals @xmath101 at @xmath87 and @xmath88 against the number of iteration steps .
the residual @xmath102 drops below @xmath103 mm for iteration with feedback control ( @xmath104 and @xmath105 ) , whereas the same remains above @xmath0 cm level with @xmath45 . shows the variation of residual percentiles @xmath101 with parameter @xmath42 with minima occurring at @xmath105 .
axial coronal sagittal axial coronal sagittal + @xmath91 + @xmath106 + ( a ) ( b ) axial coronal sagittal + @xmath91 + @xmath106
we have introduced an iterative method for numerical dvf inversion with a feedback control mechanism to improve convergence behavior .
we have described and demonstrated the key principle for altering and controlling error propagation , and a methodology for determining the control parameters numerically as well as analytically .
experimental results are in agreement with the presented error analysis and control design objectives .
we give below some additional comments on computational issues and extended feedback control mechanisms .
the iteration ( [ eq : fpim - general ] ) can be described alternatively as the following procedure with two sub - steps per iteration , [ eq : two - step - relaxed ] @xmath107 the first sub - step @xmath108 is the same as chen s iteration .
the second sub - step makes a modification by a combination of the estimates at step @xmath32 and sub - step @xmath109 .
the alternative computation procedure is mathematically equivalent with respect to convergence analysis , to iteration ( [ eq : fpim - general ] ) , although the residual feedback is not used explicitly .
the residuals can be effectively utilized in termination criteria .
commonly used termination criteria include a threshold on the numerical difference between two successive iterates and an upper bound on the total number of iteration steps .
the threshold is set based on an acceptable precision in computed results , and it should not be set higher than the computational precision used , such as ieee single or double precision . a process is terminated by the numerical threshold criterion either because the iteration has converged numerically , or because its progression is too slow .
we would need an additional indicator to differentiate the two cases . recall that the residual is related , but not necessarily equal , to the difference between successive iterates .
if the residual is not sufficiently small , the estimate is far from meeting the inverse consistency condition and the iteration has not converged well enough .
the single - parameter feedback control described in [ subsec : feedback - control ] is simple .
the parameter @xmath42 is kept constant over the spatial domain and across all iteration steps .
while this simple control mechanism proved effective with the analytic and xcat dvf data used in this study , the robustness of the convergence can be further improved by extending the control in the following two approaches .
the first extension is to turn the constant parameter to a spatially variant control function @xmath110 .
in fact , by the error propagation analysis specific to the analytic dvf ( [ eq:2d - analytic - dvfs ] ) , an ideal control scheme is spatially - variant , @xmath111 .
both analysis and numerical results ( not presented in this paper ) show that the iteration with @xmath112 takes only one or two steps for the residuals to drop to numerical zero , as determined by the machine precision .
since the residual is vector - valued in general , a spatially - variant control function @xmath36 with @xmath113 matrix local to each @xmath9 can be used .
the second extension is to make the control non - stationary , i.e. , to allow changes throughout the iteration process in adaptation to changes in the iterates or residuals .
this work was supported by the national institutes of health grant no .
the authors thank dr .
paul segars and wendy harris for their insightful suggestions on generating the xcat phantom .
they thank animesh srivastava for his generous time and effort on technical assistance .
[ sec : conflict - interest ] the authors have no relevant conflicts of interest to disclose .
we give the derivation of error propagation equations presented in this paper .
denote the error in estimating the dvf inverse at iteration @xmath32 by @xmath114 .
@xmath115 for the analytical phantom of ( [ eq:2d - analytic - dvfs ] ) , @xmath116 which gives the specific error propagation equation ( [ eq : error - propagation - customized ] ) , @xmath117 @xmath118 for the analytical phantom of ( [ eq:2d - analytic - dvfs ] ) , @xmath119 which gives the specific error propagation equation ( [ eq : error - propagation - customized ] ) , @xmath120 @xmath115 for the analytical phantom of ( [ eq:2d - analytic - dvfs ] ) , @xmath116 which gives the specific error propagation equation ( [ eq : error - propagation - customized ] ) , @xmath117 @xmath118 for the analytical phantom of ( [ eq:2d - analytic - dvfs ] ) , @xmath119 which gives the specific error propagation equation ( [ eq : error - propagation - customized ] ) , @xmath120
for general dvf , using first order taylor expansion , @xmath121 where @xmath122 .
this gives the error propagation equation of ( [ eq : error - propagation ] ) , @xmath123
the necessary and sufficient condition of the convergence of iteration process ( [ eq : fpim - single - parameter ] ) is , @xmath124
we derive the optimal @xmath42 in convergence rate for the analytical phantom ( [ eq:2d - analytic - dvfs ] ) .
@xmath125}{\max } \quad { \mathopen{}\mathclose\bgroup\leftorig } ( 1 \!- ( 1\!-\mu ) \frac{1}{1 + b\cos ( m \theta ) } { \aftergroup\egroup\rightorig})^{2}\nonumber\\ & = \displaystyle \underset{\mu}{\arg \min } \quad \underset{y \in [ -b , b]}{\max } \quad { \mathopen{}\mathclose\bgroup\leftorig } ( 1 - \frac{1 - \mu}{1 + y } { \aftergroup\egroup\rightorig})^{2}\nonumber\\ & = \displaystyle \underset{\mu}{\arg \min } \quad \begin{cases } { \mathopen{}\mathclose\bgroup\leftorig}(\frac{\mu - b}{1 - b } { \aftergroup\egroup\rightorig})^{2 } , \mu \leq b^2\\ { \mathopen{}\mathclose\bgroup\leftorig}(\frac{\mu + b}{1 + b } { \aftergroup\egroup\rightorig})^{2 } , \mu > b^2 \end{cases}\nonumber\\ & = \displaystyle b^2\end{aligned}\ ] ] a. leow , s .- c .
huang , a. geng , j. becker , s. davis , a. toga , and p. thompson .
inverse consistent mapping in 3d deformable image registration : its construction and statistical properties . in _ information processing in medical imaging _
, pages 493503 .
springer , 2005 .
w. lu , g. h. olivera , q. chen , k. j. ruchala , j. haimerl , s. l. meeks , k. m. langen , and p. a. kupelian .
deformable registration of the planning image ( kvct ) and the daily images ( mvct ) for adaptive radiation therapythis work was in part presented at the aapm meeting in seattle , july 2005 .
, 51(17):4357 , 2006 .
y. seppenwoolde , h. shirato , k. kitamura , s. shimizu , m. van herk , j. v. lebesque , and k. miyasaka .
precise and real - time measurement of 3d tumor motion in lung due to breathing and heartbeat , measured during radiotherapy .
, 53(4):822834 , 2002 . | * purpose : * the inverse of a deformation vector field ( dvf ) is often needed in deformable registration , 4d image reconstruction , and adaptive radiation therapy .
this study aims at improving both the accuracy with respect to inverse consistency and efficiency of the numerical dvf inversion by developing a fixed - point iteration method with feedback control . + * method : * we introduce an iterative method with active feedback control for dvf inversion .
the method is built upon a previous fixed - point iteration method , which is represented as a particular passive instance in the new method . at each iteration step , we measure the inconsistency , namely the residual , between the iterative inverse estimate and the input dvf . the residual is modulated by a feedback control mechanism before being incorporated into the next iterate .
the feedback control design is based on analysis of error propagation in the iteration process .
the control design goal is to suppress estimation error progressively to make the convergence region as large as possible , and to make estimate errors vanish faster whenever possible .
we demonstrated the feedback control with a single - parameter control mechanism .
the optimal parameter value is determined either analytically by a closed - form expression for analytical test data , or numerically for experimental data . the feedback control design is demonstrated and assessed with two data sets : an analytic dvf pair , and a dvf generated between two phases of the 4d extended cardiac - torso ( xcat ) digital anthropomorphic phantom .
* results : * the single - parameter feedback control improved both the convergence region and convergence rate of the iterative algorithm , for both datasets . with the analytic data ,
the iteration becomes convergent over the entire image domain , and the convergence is sped up substantially compared to the precursor method , which suffers from slow convergence or even divergence , as the deformation becomes larger . with the xcat dvf data , the new iteration method substantially outperforms the precursor method in both accuracy and efficiency ; feedback control reduced the 95th percentile of residual errors from @xmath0 mm to @xmath1 mm .
additionally , convergence rate was accelerated by at least a factor of @xmath2 for both datasets .
* conclusion : * the introduced iteration method for dvf inversion shows the previously unexplored possibility in exercising active feedback control in dvf inversion , and the unexploited potential in improving both numerical accuracy and computational efficiency .
= 1 mailto:[[`[`]author to whom correspondence should be addressed . electronic mail : ] [email protected] mailto:[[`[`]author to whom correspondence should be addressed . electronic mail : ] [email protected] |
two - dimensional ( 2d ) systems play a central role in contemporary condensed matter physics .
novel phenomena such as the quantum hall effect @xcite observed when a 2d electron gas at low temperature is subjected to a strong magnetic field , as well as practical developments based on quasi-2d systems , e.g. , high - mobility field - effect transistors @xcite or semiconductor quantum - well lasers @xcite have brought significant technological advances .
such devices are based on the quasiequilibrium response of the internal electron or electron - hole plasmas to an external stimulation .
hence it is essential to understand the fundamental quantum - statistical properties of such two - dimensional interacting plasmas at finite temperatures .
one of the well - known differences between 2d and 3d nonrelativistic quantum mechanics is the presence of at least one bound state for any symmetric attractive potential in two dimensions .
this bound state , with binding energy @xmath0 , brings a non - vanishing contribution of @xmath1 to the two - body part of the partition function even if the interaction is weak and the state is very shallow . for a dilute gas
this contribution introduces a deviation from the ideal gas law that is larger than the correction due to the fermi or bose statistics of the particles .
however , it is clear that 2d gases with a vanishing inter - particle interaction strength should be well described by free fermi or bose gas models .
this contradiction will be resolved in the present paper .
a related question is what happens to the partition function when additional bound states appear with increasing strength of interaction between the particles . in the 3d case
the answer to this question is based on a careful consideration of states in the continuum , which are modified by the interaction , i.e. , taking scattering into account in the partition function calculation @xcite . the same approach should be applied in two dimensions , however , scattering theory in two dimensions is relatively undeveloped compared to the 3d case .
for example , the relation between low - energy scattering and bound states , which has important consequences in the statistical mechanics , has been considered only very recently @xcite in two dimensions . in the present paper the connection between scattering and the statistical mechanics of a 2d plasma is studied . in the next section
we introduce the relation of the two - dimensional scattering phase shift to the partition function and show that a proper consideration of the scattering states removes discontinuities in the partition function in an analogous fashion to the 3d case . in sects .
[ short - range ] and [ ehplasma ] we explore this in more detail using an analytical model with an attractive square well potential before turning to a more realistic model of the statically screened electron - hole plasma , which is the main focus of this paper .
we also compare in both cases the influence on the second virial coefficient of the exchange interaction and the screened direct interaction .
such considerations are crucial in understanding the nature of the strongly correlated electron - hole plasma in semiconductor quantum wells .
the two - body interaction part of the partition function of 2d interacting boltzmann particles is given by @xmath2 where @xmath3,@xmath4 is the reduced mass , @xmath5 is the projection of the angular momentum onto the axis normal to the plane of 2d motion ( @xmath6 ) , @xmath7 is the 2d scattering phase shift @xcite dependent on the relative - motion momentum @xmath8 , @xmath9 are the bound - state energies ( index @xmath10 enumerates bound states with given @xmath11 ) , and the double sum in the first term ranges only over bound states .
( [ eq1 ] ) is the 2d analogue of the beth - uhlenbeck formula and can be derived in the same fashion as in the 3d case @xcite .
often only the first term in eq .
( [ eq1 ] ) is considered when calculating an internal partition function , neglecting the phase shift term . for an attractive potential @xmath12 , as @xmath13 decreases bound state energies increase towards the continuum .
as such a state reaches zero energy , a partition function that contains only the bound state sum will be discontinuous .
these unphysical singularities would extend to all the thermal properties , such as pressure and specific heat .
integrating by parts we can rewrite eq .
( [ eq1 ] ) as @xmath14 for nonzero temperature the integral term in the right - hand side of eq .
( [ eq2 ] ) is a smooth function of the interaction strength @xmath15 . if the phase shifts satisfy the condition : @xmath16 where @xmath17 is the number of bound states with given @xmath5 , the zero - energy part of the phase shift integral in eq .
( [ eq1 ] ) exactly cancels the zero - energy part of the bound state sum , removing the discontinuity in @xmath18 as a function of the interaction strength .
this cancellation is similar to the well known behaviour in 3d where the partition function discontinuities are removed @xcite with the help of levinson s theorem @xcite .
equation ( [ eq3 ] ) constitutes the 2d statement of levinson s theorem . as a central theorem of scattering theory @xcite ,
levinson s theorem has been discussed for dirac particles , multichannel scattering , multi - particle single - channel scattering , one - dimensional scattering systems , impurities in aharonov - bohm rings , systems with non - uniform effective mass , and even for time - periodic potentials @xcite .
however , its applicability to the 2d scattering problem has been considered only recently . in ref .
@xcite the 2d statement of levinson s theorem , eq .
( [ eq3 ] ) , was proposed and verified empirically , while in ref .
@xcite this theorem was more rigorously established for cutoff potentials using the green - function method .
the above arguments , based on the partition function continuity requirement , provide an additional statistical mechanical justification of levinson s theorem in 2d .
the two - body interaction part of the partition function can be used to calculate the second virial coefficient , @xmath19 , which characterizes the first correction to the ideal gas law in the low - density expansion of pressure @xmath20 @xmath21 is positive for repulsive potentials , causing an increase in the pressure over its ideal - gas value , and negative for attractive potentials , causing a decrease in the pressure .
a calculation of the second virial coefficient is meaningful in the dilute gas regime , where the mean inter - particle spacing @xmath22 is larger than the thermal wavelength @xmath23 and the higher order terms in eq .
( [ eq4 ] ) are negligible . for
a free , bose or fermi gas in two dimensions , @xmath24 @xcite , the plus sign applying to the fermi case , as the pauli principle introduces an effective repulsion between fermions , thereby increasing the pressure .
there is extensive ongoing research in the statistical mechanics of anyons , 2d particles obeying fractional statistics @xcite , and the second virial coefficient of a free anyon gas lies in between the bosonic and fermionic value @xcite . in this paper
we focus on the relative importance of interaction - induced bound and scattering states on the second virial coefficient .
the 3d analysis of the second virial coefficient @xcite is easily reformulated for a 2d interacting gas @xcite . for a system of identical particles with spin @xmath25
the second virial coefficient is @xmath26 where the upper sign is for bosons and the lower sign for fermions .
the exclusion principle modifies the sum over @xmath5 in eq .
( [ eq2 ] ) depending on the angular momentum parity , and the partition function ( using eq .
( [ eq3 ] ) ) is @xmath27 \nonumber \\ & & \nonumber\\ & ~+~&(2s+1){2\over{\pi q_t^2 } } \int_0^\infty q dq \left[\sum_{m=-\infty}^{\infty } \left(s+{1\pm(-1)^m \over 2}\right ) \delta_m(q ) \right ] \exp(-q^2/q_t^2)~. \label{eq6 } \end{aligned}\ ] ] the electron - hole plasma constitutes a mixture of two components and for a binary mixture of components @xmath28 and @xmath29 having second virial coefficients @xmath30 and @xmath31 and densities @xmath32 and @xmath33 , respectively , the second virial coefficient is @xcite @xmath34 in eq .
( [ eq7 ] ) @xmath35 and @xmath36 where @xmath37 and @xmath18 is given by eq .
( [ eq1 ] ) or eqs .
( [ eq2 ] ) and ( [ eq3 ] ) with the properly chosen reduced mass @xmath38 .
our first example , the boltzmann gas with weak short - range attraction , is chosen to elucidate how the second virial coefficient at given temperature vanishes when the binding energy @xmath39 decreases , even though the bound - state part of the internal partition function @xmath40 approaches unity rather than vanishes .
to trace the precise nature of the cancellation of @xmath41 it is convenient to use eq .
( [ eq1 ] ) for @xmath18 without the application of levinson s theorem .
let us assume that the 2d particles interact via an attractive square - well potential of radius @xmath42 and depth @xmath43 .
this simple model allows analytical treatment which provides insight into the generic behaviour of a gas of attracting particles . to evaluate the partition function
we need to analyse both the bound and scattering states in this potential .
the binding energies for any value of angular momentum @xmath5 can be easily found by matching the logarithmic derivative of the radial wave function at @xmath44 : @xmath45 where @xmath46 , @xmath47 , @xmath48 is the bessel function of the first kind and @xmath49 is the modified bessel function of the second kind .
note that for @xmath50(i.e .
smaller than the first root of @xmath51 ) there is only one bound state ( having @xmath52 ) and for @xmath53 this state is very shallow , e.g. for @xmath54 , @xmath55 , for @xmath56 , @xmath57 . for a shallow @xmath58 state
the transcendental equation for binding energies eq .
( [ eq10 ] ) reduces to @xmath59 where @xmath60 ( @xmath61 is euler s constant ) . for the unbound states with positive energy of the relative motion , @xmath62 , scattering phase shifts can be found in a similar fashion . for small values of the momentum , @xmath63 , all phase shifts for @xmath64
are small compared to @xmath65 ( @xmath25-wave scattering ) @xcite .
the tangent of the @xmath25-wave scattering phase shift for @xmath66 ( i.e. @xmath67 ) is given by @xcite : @xmath68 substituting @xmath69 from eq .
( [ eq11 ] ) into eq .
( [ eq12 ] ) we get @xmath70 note that this expression does not contain parameters of the potential @xmath71 and @xmath72 explicitly , and it is valid for an arbitrary potential well with a shallow @xmath73 level @xcite . since the integrand in the partition function , eq .
( [ eq1 ] ) , contains an exponential factor @xmath74 , the wavevectors @xmath75 which are larger than the thermal wave vector @xmath76 give negligible contribution to the value of the integral .
therefore for the short - range interaction or for low temperature , satisfying condition @xmath77 , the scattering phase shifts need only be considered for @xmath78
. then all the terms in the phase shift sum in eq .
( [ eq1 ] ) can be neglected except for the term with @xmath52 .
finding the derivative of @xmath65 from eq .
( [ eq13 ] ) we obtain for the two - body interaction part of the partition function @xmath79 the integral in eq .
( [ eq15 ] ) is the ramanujan integral @xcite which can be rewritten as @xcite @xmath80 where @xmath81 with @xmath82 .
thus , the partition function acquires a very simple form @xmath83 a similar result has been obtained recently for contact - interacting particles @xcite . to consider the small @xmath84 asymptotic of the function @xmath85 it is convenient to expand the integral eq .
( [ eq17 ] ) in descending powers of @xmath86 : @xmath87^{-3}\right)~. \label{eq19}\ ] ] from eq .
( [ eq19 ] ) one can see that @xmath18 and hence the second virial coefficient @xmath88 both vanish when @xmath89 , although one bound state always exists .
so the lowest - order density correction to the 2d ideal gas law vanishes only slowly as @xmath90 as the binding energy is reduced .
note that , when the potential supports several bound states , the contribution of any shallow bound state with @xmath58 is cancelled by the scattering phase shift integral in the `` logarithmic '' manner described above . for @xmath64
the cancellation has a power - law dependence in @xmath91 @xcite .
this implies that higher - order levinson s theorems responsible for continuity of the partition function derivatives @xcite are different for @xmath58 and @xmath92 , whereas the zeroth - order levinson s theorem in two dimensions has the same form , eq .
( [ eq3 ] ) , for all @xmath11 . for extremely weak interaction potential , such that @xmath93 , from eqs .
( [ eq11 ] ) and ( [ eq19 ] ) it follows that @xmath94 , which coincides with the perturbation theory result . in the other limit for large @xmath84 values ( @xmath95 ) , @xmath96 @xcite , therefore the exponential dependence of the partition function on the binding energy
is recovered . in fig .
[ fig1 ] we plot the ratio of the total partition function @xmath97 to its bound - state part @xmath98 as a function of @xmath99 .
we do this for both the full expression , eq .
( [ eq17 ] ) , and the first two terms in the asymptotic expansion , eq .
( [ eq19 ] ) .
one can see that the asymptotic expression ( dashed line in fig .
[ fig1 ] ) is accurate only for very small values of @xmath99 . over a wide range of @xmath91 both scattering and bound - state terms
are important , e.g. when @xmath100 the scattering term produces a 20% correction to @xmath97 .
when @xmath101 the bound - state contribution dominates completely .
16truecm it is instructive to compare the contributions to the second virial coefficient of the direct and the exchange interactions .
for example for a gas of spinless bosons the second virial coefficient is @xmath102 . for @xmath18 given by eq .
( [ eq18 ] ) , the direct interaction term ( @xmath103 ) is smaller than the exchange term ( @xmath104 ) for small binding energies .
thus , as required for vanishing binding energy , the second virial coefficient is governed by the statistics of particles , despite the existence of a single bound state .
in this section we study a more realistic model of an interacting gas than the one considered in sect .
[ short - range ] .
we consider a mixture of the two types of fermi particles , positively charged holes ( h ) and negatively charged electrons ( e ) .
this model is important for understanding the thermodynamic and optical properties of the electron - hole plasma in semiconductor quantum wells . in the low - temperature , low - density limit most electron - hole pairs
are bound into excitons .
this limit has been studied extensively because of the recent proposals of exciton condensates in this system @xcite .
the properties of the degenerate 2d electron - hole plasma are also well known @xcite .
we consider the case when the temperature is comparable to the exciton binding energy so the occupation of continuum states ( and therefore screening ) is significant , although the carrier density is low enough to neglect the terms higher than @xmath105 in the virial expansion .
the temperature and density conditions under investigation are close to those in the regime where excitonic gain in wide - gap semiconductors is anticipated @xcite .
we assume for simplicity equal masses @xmath106 and spins @xmath107 for both species .
then @xmath108 and the second virial coefficient for the mixture eq . ( [ eq7 ] ) acquires the form : @xmath109 here @xmath110 and the charge neutrality condition @xmath111 is taken into account .
the screened coulomb attraction between electrons and holes is modeled using the fourier representation of the interaction potential : @xmath112 where @xmath113 is the screening wave vector .
hereafter we employ 3d excitonic rydberg units where length and energy are scaled , respectively , by the effective bohr radius @xmath114 and rydberg ry@xmath115 . for electron - electron and hole - hole repulsion the same potential with the opposite sign
equation ( [ eq21 ] ) is the well - known thomas - fermi expression for the coulomb potential statically screened by a 2d electron gas .
being the long - wavelength static limit of the random phase approximation , eq .
( [ eq21 ] ) is a simple model for the screened coulomb potential in two dimensions .
nevertheless , this expression reflects the fact that the statically screened potential in two dimensions decreases at large distances slower than in the 3d case ( as a power law rather than exponentially ) . despite numerous more realistic corrections
@xcite eq . ( [ eq21 ] ) remains the most widely used approximation for the 2d screening @xcite .
this potential has been known for more than three decades @xcite but some of its unusual properties were only discovered recently , e.g. the existence of a remarkably simple relation between @xmath113 and the number of bound states .
namely , with decreasing screening , new bound states appear at the critical integer values of the screening length given by the simple formula@xcite @xmath116 where @xmath5 is the angular momentum and @xmath117 indicates the number of nonzero nodes of the radial wave function .
equation ( [ eq22 ] ) shows that several bound states corresponding to the given value of @xmath118 appear simultaneously .
this degeneracy is different from the degeneracy for the bound states of the unscreened 2d exciton ( or hydrogen atom ) , for which the states with the same value of @xmath119 are degenerate @xcite .
the hidden symmetry that underlines this new degeneracy has not been fully understood yet , and eq .
( [ eq22 ] ) still lacks a rigorous analytical derivation .
we also consider the low - density ( nondegenerate ) limit , when there is no pauli blocking and the self - energy correction @xcite to the beth - uhlenbeck formula can be neglected and eqs .
( [ eq2],[eq3 ] ) and eq .
( [ eq6 ] ) can be used for @xmath120 and @xmath121 respectively .
the shortcomings of this model for the quantitative description of a real system of photoexcited electrons and holes in semiconductor quantum wells are self - evident ; however , it does provide a tractable model containing all the salient features of the system . to find the second virial coefficient given by eq .
( [ eq20 ] ) one must calculate the binding energies and scattering phase shifts entering the partition functions @xmath120 and @xmath121 .
we use for this purpose the 2d modification of the variable - phase method @xcite known from scattering theory . in this method
the scattering phase shift and the function defining bound - state energies can be obtained as a large distance limit of the phase function , which satisfies the first - order , nonlinear riccati equation originating from the radial schrdinger equation .
the variable - phase method application to scattering and bound states in the screened coulomb potential ( [ eq21 ] ) is described in detail in ref .
the method is especially effective for calculation of shallow - state binding energies and low - energy scattering phase shifts .
figure [ fig2 ] shows the results from the calculation of the electron - hole part of the partition function , @xmath122 , which contains both the bound state sum and the scattering phase shift integral . in this figure @xmath120
is plotted as a function of the inverse screening wave number @xmath123 .
16truecm for the electron - hole plasma @xmath113 is a function of carrier density @xcite , and in the purely 2d case @xmath124 for low densities and is independent of density , @xmath125 for @xmath126 , in the degenerate limit .
here we treat @xmath113 as a parameter characterizing the strength of the screened interaction potential eq .
( [ eq22 ] ) . to emphasize the role of scattering we show on the same plot the bound - state sum , @xmath127 , which exhibits jumps whenever new bound states appear [ i.e. when @xmath113 satisfies eq .
( [ eq22 ] ) ] .
these jumps become higher with increasing screening length @xmath128 since several bound states appear simultaneously .
as can be seen the additional scattering phase shift contributions completely remove these jumps .
the partition function is plotted for two values of the ratio of @xmath129 to the excitonic rydberg , @xmath130 and @xmath131 , which roughly correspond to znse ( or gan ) and gaas at room temperature .
one can see that for high temperature ( or low binding energy ) the bound - state contributions to @xmath120 are suppressed by the scattering phase shift integral more strongly than in the @xmath132 case . in fig .
[ fig3 ] the second virial coefficient @xmath133 ( scaled by @xmath134 ) is plotted versus the screening wave number @xmath113 for two different values of @xmath135 .
16truecm equation ( [ eq20 ] ) is used for the calculation of @xmath136 , and the repulsion ( @xmath137 ) term partially compensates the @xmath120 term .
this compensation is especially significant in the high - temperature case @xmath138 , in which the 2d electron - hole plasma behaves much like an ideal gas over a wide range of screening wave vectors . for @xmath132
the electron - hole attraction term dominates and the plasma is strongly correlated for all values of @xmath139 . in this case a small statistical repulsion ( @xmath140 , horizontal line in fig . [ fig3 ] ) , which is due to the fermionic nature of electrons and holes , can be completely neglected .
thus , at room temperature the electron - hole plasma in gaas - based quantum wells can be treated as an ideal gas , whereas in wide - gap semiconductors ( e.g. , znse or gan ) due to the high value of ry@xmath115 the 2d electron - hole plasma is strongly correlated and excitonic effects are important for its thermodynamic properties .
in this paper we show that a proper account of scattering eliminates discontinuities in thermodynamic properties of the nonideal 2d gas whenever extra bound states appear with small increase of the strength of an attractive potential .
this treatment provides a heuristic proof of levinson s theorem in two dimensions .
we trace the way in which the bound - state contribution to the partition function of the 2d gas , with a weak short - range attraction between its particles , vanishes when the binding energy decreases . a weak @xmath90 binding energy dependence of the second virial coefficient of such a gas is found for @xmath141 .
a consistent treatment of bound and scattering states in a screened coulomb potential allows us to calculate the quantum - mechanical second virial coefficient of the dilute 2d electron - hole plasma and to establish the difference between the nearly ideal electron - hole gas in gaas and the strongly correlated exciton / free - carrier plasma in wide - gap semiconductors .
the 2d electron - hole plasma was considered in the low - density nondegenerate limit only .
transition to the strongly degenerate fermi limit and related questions of pauli blocking and self - energy corrections to the beth - uhlenbeck formula in 2d remain the subject of further research . also at a. f. ioffe physico - technical institute , st.petersburg , russia .
, edited by r. e. prange and s. m. girvin ( springer , new york , 1990 )
. t. ando , a. b. fowler , and f. stern , rev .
mod . phys . * 54 * , 437 ( 1982 ) .
w. w. chow , s. w. koch , and m. sargent iii , _ semiconductor - laser physics _
( springer , berlin , 1994 ) .
a. i. larkin , zh .
38 , 1896 ( 1960 ) [ sov .
jetp * 11 * , 1363 ( 1960 ) ] .
f. j. rogers , h. c. graboske jr , and h. e. dewitt , phys .
34a * , 127 ( 1971 ) ; f. j. rogers , phys . rev . a * 4 * , 1145 ( 1971 ) .
m. e. portnoi and i. galbraith , solid state commun . , * 103 * , 325 ( 1997 ) ; cond - mat/9705136 .
lin , phys . rev .
a * 56 * , 1938 ( 1997 ) .
f. stern and w. e. howard , phys . rev . * 163 * , 816 ( 1967 ) .
e. beth and g. e. uhlenbeck , physica ( amsterdam ) * 4 * , 915 ( 1937 ) ; see also , e.g. , l. d. landau and e. m. lifshitz , _ statistical physics _ ( pergamon , new york , 1980 ) , pt .
n. levinson , k. dan .
selsk . mat .
. medd . * 25 * , 3 ( 1949 ) .
r. g. newton , _ scattering theory of waves and particles _
( mcgraw - hill , new york , 1966 ) .
ma , phys .
76 * , 3654 ( 1996 ) ; n. poliatzky , _ ibid . _
* 76 * , 3655 ( 1996 ) ; f. vidal and j. letourneaux , phys .
c * 45 * , 418 ( 1992 ) ; l. rosenberg and l. spruch , phys .
a * 54 * , 4978 ( 1996 ) ; * 54 * , 4985 ( 1996 ) ; k. a. kiers and w. van dijk , j. math . phys .
* 37 * , 6033 ( 1996 ) ; m. s. debianchi , _ ibid .
_ * 35 * , 2719 ( 1994 ) ; v. milanovi _ et al .
a * 170 * , 127 ( 1992 ) ; v. milanovi _ et al . _ ,
a * 25 * , l1305 ( 1992 ) ; m. diventra _
et al . _ ,
b * 55 * , 1353 ( 1997 ) ; p. a. martin and m. s. debianchi , europhys . lett . * 34 * , 639 ( 1996 ) , and references therein .
r. l. siddon and m. schick , phys . rev .
a * 9 * , 907 ( 1973 ) ; j. g. dash , _ films on solid surfaces _ ( academic , new york , 1975 ) .
p. giacconi , f. matoni , and r. soldati , phys .
b * 53 * , 10065 ( 1996 ) , and references therein .
p. arovas , r. schrieffer , f. wilczek , and a. zee , nucl .
b * 251 * , 117 ( 1985 ) . for a review on anyons
see , e.g. , _ fractional statistics and anyon superconductivity _ , ed . by f. wilczek ( world scientific , singapore , 1990 ) .
j. e. kilpatrick , j. chem .
* 21 * , 274 ( 1953 ) ; j. e. kilpatrick , w. e. keller , e. f. hammel , and n. metropolis , phys .
94 * , 1103 ( 1954 ) .
l. d. landau and e. m. lifshitz , _ quantum mechanics ( non - relativistic theory ) _
( butterworth - heinemann , oxford , 1997 ) .
m. e. portnoi , pisma zh .
14 , 1252 ( 1988 ) [ sov
* 14 * , 547 ( 1988 ) ] .
it is interesting to note that the expression for the scattering cross - section for resonant @xmath25-wave scattering in two dimensions , @xmath142 $ ] , despite being quite different from its 3d equivalent @xcite , nevertheless also has a pole at @xmath143 .
g. h. hardy , _ ramanujan _ ( cambridge university press , cambridge , 1940 ) , p. 196 . i. s. gradshtein and i. m. ryzhik , _ tables of integrals , series , and products _
( academic press , new york , 1994 ) . m. e. portnoi and i. galbraith , unpublished .
d. boll , ann .
( ny ) * 121 * , 131 ( 1979 ) .
, edited by a. erdlyi ( mcgraw - hill , new york , 1955 ) , vol . 1 ,
x. zhu , p. b. littlewood , m. s. hybertsen , and t. m. rice , phys .
lett . * 74 * , 1633 ( 1995 ) .
j. fernndez - rossier and c. tejedor , phys .
lett . * 78 * , 4809 ( 1997 ) , and references therein .
s. schmitt - rink , d. s. chemla , and d. a. b. miller , adv .
phys . * 38 * , 89 ( 1989 ) .
i. galbraith , in : _ microscopic theory of semiconductors : quantum kinetics , confinement and lasers _ , ed . by s. w. koch ( world scientific , singapore , 1996 ) , p.211 , and references therein .
c. bulutay and m. tomak , phys .
b * 54 * , 14643 ( 1996 ) ; k. schrfer _
et al . _ ,
superlattices and microstructures * 21 * , 223 ( 1997 ) ; a. gold and a. ghazali , j. phys . : condens .
matter * 9 * , 6885 ( 1997 ) , and references therein .
h. haug and s. w. koch , _ quantum theory of the optical and electronic properties of semiconductors _
( world scientific , singapore , 1994 ) .
j. a. reyes , m. del castillo - mussot , phys .
b * 56 * , 14893 ( 1997 ) , and references therein .
s. flgge and h. marschall , _ rechenmethoden der quantentheorie _
( springer , berlin , 1952 ) , problem 24 .
r. zimmermann , h. stolz , phys .
status solidi b * 131 * , 151 ( 1985 ) .
f. calogero , _ variable phase approach to potential scattering _
( academic , new york , 1967 ) . | we consider scattering state contributions to the partition function of a two - dimensional ( 2d ) plasma in addition to the bound - state sum .
a partition function continuity requirement is used to provide a statistical mechanical heuristic proof of levinson s theorem in two dimensions .
we show that a proper account of scattering eliminates singularities in thermodynamic properties of the nonideal 2d gas caused by the emergence of additional bound states as the strength of an attractive potential is increased .
the bound - state contribution to the partition function of the 2d gas , with a weak short - range attraction between its particles , is found to vanish logarithmically as the binding energy decreases .
a consistent treatment of bound and scattering states in a screened coulomb potential allowed us to calculate the quantum - mechanical second virial coefficient of the dilute 2d electron - hole plasma and to establish the difference between the nearly ideal electron - hole gas in gaas and the strongly correlated exciton / free - carrier plasma in wide - gap semiconductors such as znse or gan . |
gamma - ray line astronomy has opened a new and unique window for studying nucleosynthesis in our galaxy .
the singular advantage of gamma - ray spectroscopy over other observations is that it offers the opportunity to detect directly and identify uniquely individual isotopes at their birthplaces .
the rate at which radioactive decays proceed is in general unaffected by the physical conditions in their environment , such as temperature or density .
the interstellar medium is not dense enough to attenuate gamma rays , so that radioactive decays can be observed throughout our galaxy .
recent reviews on implications of gamma - ray observations for nucleosynthesis in our galaxy can be found in @xcite and @xcite .
the nucleosynthesis results presented in the following have all been obtained from observations with the two main instruments on board the integral observatory : the spectrometer spi and the imager ibis ( for details regarding the instruments , see @xcite and references therein ) .
these two instruments are complementary in their characteristics , providing an unprecedented view of the universe at hard x - ray and soft gamma - ray energies .
the imaging ge spectrometer spi offers high spectral resolution of about 2.1 kev fwhm at 511 kev combined for the first time with moderate spatial resolution ( fwhm about @xmath3 ) .
the imager ibis offers excellent spatial resolution of about @xmath4 fwhm at moderate spectral resolution ( fwhm about 38 kev at 511 kev ) .
the radioisotope @xmath2ti is primarily produced in the so - called @xmath5-rich freeze - out of material initially in nuclear statistical equilibrium .
the main site for @xmath5-rich freeze - out to occur is thought to be the innermost layers of core - collapse supernovae ( ccsne ) , although sub - chandrasekhar mass white dwarf type ia sne have also been proposed ( ( * ? ? ?
* woosley & weaver 1994 ) ) .
ni and @xmath2ti inside a 20 m@xmath6 star after the passage of the shock front .
right panel : radial abundance profiles ( mass fractions ) of @xmath0al and @xmath1fe inside a 25 m@xmath6 star after the passage of the shock front .
both figures were adapted from @xcite.,title="fig:",width=264 ] ni and @xmath2ti inside a 20 m@xmath6 star after the passage of the shock front .
right panel : radial abundance profiles ( mass fractions ) of @xmath0al and @xmath1fe inside a 25 m@xmath6 star after the passage of the shock front .
both figures were adapted from @xcite.,title="fig:",width=264 ] the @xmath2ti yield of ccsne is notoriously difficult to calculate because it depends sensitively on the so - called mass cut , the explosion energy , and the ( a)symmetry of the explosion .
the mass cut , which has not yet been successfully calculated and is illustrated in the left panel of fig .
[ prantzos_44tifig ] , is the notional surface separating material that is ejected from material that will fall back onto the compact remnant ( neutron star or black hole ) of the explosion .
@xmath2ti is believed to be produced in the deepest layers of the exploding star that may be ejected , depending on the precise location of the mass cut .
the amount of synthesized @xmath2ti also depends sensitively on the explosion energy and ( a)symmetry .
theoretical calculations indicate that both increased explosion energy and increased asymmetry result in an increased @xmath2ti yield .
observationally , the presence of the radioisotope @xmath2ti is revealed to the gamma - ray astronomer through the emission of three gamma - ray lines .
the decay @xmath2ti
@xmath7 @xmath2sc ( @xmath8 y ) gives rise to gamma rays at 67.9 kev and 78.4 kev ; the subsequent decay @xmath2sc @xmath7 @xmath2ca ( @xmath9 h ) gives rise to a line at 1157.0 kev .
the astrophysical interest in @xmath2ti is two - fold . clearly , the amount and the velocity of @xmath2ti is a very powerful probe of the explosion mechanism and dynamics of ccsne , which are still poorly understood .
in addition , the @xmath2ti gamma - ray line emission is an ideal indicator of young sn remnants ( snrs ) .
the lifetime is about 90 y , which roughly coincides with the expected recurrence time interval for ccsne in our galaxy .
it is therefore expected that with a sufficiently sensitive instrument a few young snrs should be visible in our galaxy at the current epoch .
the most sensitive search to date for young snrs at gamma - ray energies was performed by @xcite who used the first year of integral observations to search for 68 kev and 78 kev line emission in the inner galaxy with the imager ibis .
this search addresses a long - standing puzzle linking the galactic sn rate and galactic chemical evolution : given current estimates of the present - day rates of thermonuclear and ccsne and their yields , these events can only account for about @xmath10 of the solar @xmath2ca abundance based on chemical evolution models and assuming that all @xmath2ca is formed as @xmath2ti ( ( * ? ? ?
* leising & share 1994 ) ) .
at the same time , given these sn properties , combined with models for their galactic distribution , past missions should have detected a few young snrs even with their lower sensitivities and detections were certainly expected for the unprecedented limiting point source sensitivity achieved with ibis . however , as was the case in less sensitive previous searches , none have been found . to assess the implications of the non - detection of young snrs , @xcite estimated the probability that at least one @xmath2ti point source is detectable by generating monte carlo distributions using current estimates of the rates of thermonuclear and ccsne , of their yields , and of their galactic distribution ; sn explosions were simulated as a function of the recurrence time and the @xmath2ti yield of type ii sne .
a typical result is depicted in fig .
[ renaud_fig ] .
the monte carlo study rules out two obvious solutions to the @xmath2ca puzzle .
the estimated galactic @xmath2ti production rate of ccsne could in principle be increased by postulating that the galactic sn rate has been underestimated , or by postulating that the @xmath2ti yields have been underestimated .
neither solution is viable , because in order to account for the solar @xmath2ca abundance by sne a present - day production rate of @xmath11 m@xmath6 century@xmath12 is required which is excluded at a greater than 95% confidence level for any acceptable combination of sn rates and @xmath2ti yields .
a third solution , compatible with existing data , is the existence of another , rare but high yield , galactic @xmath2ti source such as sub - chandrasekhar mass white dwarf type ia sne .
the latest search for young snrs with ibis therefore highlights important open questions concerning the chemical evolution of our galaxy , the production of @xmath2ti in sne , and the galactic sn rate ( ( * ? ? ?
* renaud 2004 ) ) .
@xmath0al and @xmath1fe are produced hydrostatically as well as explosively in massive stars .
the production of @xmath0al requires free protons , hence it is mainly produced in the h , ne , and c burning layers ; the production of @xmath1fe requires free neutrons , hence it is mainly produced in c , ne , and he burning layers , as indicated in the right panel of fig .
[ prantzos_44tifig ] .
products of hydrostatic nucleosynthesis can escape the massive star through stellar winds , in particular during the wolf - rayet ( wr ) phase of the most massive stars .
the @xmath0al , being produced in h burning layers near the surface , can escape most easily .
when the star explodes in a ccsn , about equal amounts of @xmath0al and @xmath1fe are ejected .
the yields of @xmath1fe and in particular of @xmath0al are hard to predict because of the important role of stellar winds , which in turn depend e.g. on the metallicity of the star or its rotation . both radioisotopes can be observed by gamma - ray line spectroscopy .
the decay of @xmath0al ( @xmath13 y ) gives rise to a single gamma - ray line at 1808.6 kev .
the decay @xmath1fe
@xmath7 @xmath1co
( @xmath14 y ) is revealed by a line at 58.6 kev , the subsequent decay @xmath1co @xmath7 @xmath1ni ( @xmath15 y ) is revealed by two lines at 1173.2 kev and 1332.5 kev .
astrophysically , the subtle differences in the production zones of the radioisotopes @xmath0al and @xmath1fe render their combined study an important probe of massive star nucleosynthesis .
furthermore , since the lifetimes of these two radioisotopes are much shorter than galactic evolution timescales but are large enough to accumulate in the ism , @xmath0al and @xmath1fe are ideally suited to study the sites and the distribution of recent galactic star formation .
the sky distribution of galactic @xmath0al was first mapped using the comptel instrument ; it was found to be very similar to that of massive stars in our galaxy , establishing the massive star origin of @xmath0al ( ( * ? ? ?
* kndlseder 1999 ) ) . however , the relative importance of wr star winds and of ccsne for seeding the ism remained unclear .
early high - resolution spectroscopy with the balloon borne gris instrument by @xcite yielded an unexpectedly large line width of about 5.4 kev fwhm and complicated rather than clarified the origin of @xmath0al as such substantial line broadening could only be explained by rather unusual circumstances ( ( * ? ? ?
* sturner & naya 1999 ) ) .
spatially resolved spectroscopy , as afforded by the imaging spectrometer spi , is expected to resolve the issue , as it will make it possible to probe the imprint of the dynamics of the ism local to the @xmath0al sources and that of galactic differential rotation on shape and centroid of the 1808.6 kev line ( modelled e.g.by ( * ? ? ?
* kretschmer 2003 ) ) . using the first 1.5 years of observations with spi , (
* ; * ? ? ?
* diehl ( 2005a , b ) ) obtained a spectrum of the 1808.6 kev line from the inner galaxy of unprecedented quality ( see left panel of fig .
[ 26al_60fe_spectra ] ) .
the line centroid is unshifted ; the line width is intrinsically narrow with a fhwm of about 1.2 kev ( and a 2@xmath16 upper limit of 2.8 kev ) , ruling out the earlier gris result .
these line parameters are consistent with expectations from the @xcite model , which for the inner galaxy predicts no line shift and a line width of about 1 kev due to galactic differential rotation ( leaving room for a small amount of additional broadening due to modest ism turbulence around the @xmath0al sources ) .
the @xmath0al line is detected with such high significance by spi that spatially resolved spectroscopy in the inner galaxy can be performed for the first time , with important implications for the amount and origin of @xmath0al , and the galactic star formation and sn rates as discussed by @xcite .
al gamma - ray line emission from the inner galaxy ( data points ) and a gaussian line fit ( solid line ) as obtained by @xcite .
right panel : the overlayed and summed spectrum of the observed @xmath1fe lines at 1173.2 kev and 1332.5 kev ( data points ) and a gaussian line fit ( dashed line ) according to @xcite.,title="fig:",width=264 ] al gamma - ray line emission from the inner galaxy ( data points ) and a gaussian line fit ( solid line ) as obtained by @xcite .
right panel : the overlayed and summed spectrum of the observed @xmath1fe lines at 1173.2 kev and 1332.5 kev ( data points ) and a gaussian line fit ( dashed line ) according to @xcite.,title="fig:",width=264 ] one of the great successes for spi was the first significant measurement of @xmath1fe emission from our galaxy by @xcite using the observations of the first year .
the summed and overlayed spectrum of the two observed lines is depicted in the right panel of fig .
[ 26al_60fe_spectra ] .
the observed flux per line is @xmath17 ph @xmath18 s@xmath12 , which translates into a line flux ratio of @xmath1fe to @xmath0al of @xmath19 per @xmath1fe line .
the detection of @xmath1fe has important implications for the nucleosynthesis in massive stars , as was first pointed out by ( * ? ? ?
* ; * ? ? ?
* prantzos ( 2004 , 2005 ) ) .
he integrated the predicted yields for type ii sne over an initial mass function and found that if he only considered the lower mass range , where the wr phase is absent or weak , the @xmath1fe to @xmath0al line flux ratio is significantly overpredicted .
this means that there must be an additional @xmath0al source ; type ii sne do not produce enough @xmath0al .
this additional source could be wr stars , because when the integration is extended up to the highest masses , taking into account the @xmath0al expelled in the wr winds , the observed @xmath1fe to @xmath0al line flux ratio can be reproduced .
however , one should keep in mind that wr star yields are still uncertain , as they depend sensitively on e.g. metallicity and rotation .
the annihilation of positrons with electrons gives rise to two characteristic emissions at gamma - ray energies : the hallmark line at 511 kev , and the unique three - photon positronium ( ps ) continuum emission ( cf .
* guessoum , jean , & gillard 2005 ) ) .
the detailed shape of the 511 kev annihilation line , as well as the positronium fraction ( the fraction of positrons that annihilate through ps formation ) , depend on the physical properties of the annihilation media ; therefore detailed spectroscopy of the positron annihilation can provide unique information on the annihilation media and processes .
positrons can be produced through a variety of processes and by a variety of objects in our galaxy .
the main production processes are @xmath20 decay of radioactive nuclei ( formed in sne , novae , hypernovae , wr stars , and agb stars ) , the decay of @xmath21s resulting from interactions of energetic nuclei ( in the ism ) , pair ( electron - positron ) production in @xmath22-@xmath22 interactions ( close to compact objects and in gamma - ray bursts ) , and pair production by electrons in strong magnetic fields ( around pulsars ) .
exotic processes , such as the annihilation or decay of light dark matter , have also been proposed ( ( * ? ? ?
* boehm 2004 ) ) . in spite of this well developed understanding of the fundamental physics of positron production , theoretical models for the positron yield of the various sources are still highly uncertain due to uncertainties concerning the astrophysical conditions in the different sources , and their galactic distribution and duty cycle .
considering also the limited quality of existing observations , it is not surprising that the origin of the positrons is still poorly understood . at this stage
improved measurements of the spatial distribution and the spectrum of the galactic annihilation radiation are crucial for constraining better the many models .
boxcar average .
further details are given in the text.,width=302 ] investigations of the sky distribution of the annihilation radiation promise to provide clues to the identification of the source(s ) of positrons in our galaxy .
first maps of the annihilation radiation , limited to the inner regions of our galaxy , were obtained using the osse instrument on board the compton gamma - ray observatory in the 511 kev line and in ps continuum emission ( e.g. ( * ? ? ?
* purcel 1997 ) , ( * ? ? ?
* milne 2001 ) ) . improved mapping of the 511 kev line is feasible with the commissioning of spi . during the first year of the mission
most of the celestial sphere was observed . using these observations , first sky maps in the 511 kev positron annihilation line ( (
* kndlseder 2005 ) ) and in the ps continuum emission ( ( * ? ? ?
* weidenspointner 2005 ) ) have been obtained with the spi spectrometer ( fig .
[ 511_poscnt_fig ] , top and bottom ) . in both maps of positron annihilation radiation
, the only prominent signal seen is that from the galactic bulge region .
the surface brightness of any emission from any other sky regions , in particular from the galactic disk , is much fainter .
the emission appears to be symmetric about the galactic center ( gc ) , and its centroid coincides well with the gc .
the differences between the two maps , as well as any apparent small scale sub - structure in the maps , are not significant , as will be demonstrated below . a more quantitative approach for studying the galactic distribution of the observed extended emission is model fitting . from modeling the positron annihilation radiation by an ellipsoidal distribution with a gaussian radial profile
it can be concluded that the annihilation radiation is spherically symmetric with a fwhm of about @xmath23 and centered at the gc ( ( * ? ? ?
* kndlseder 2005 ) , ( * ? ? ? * weidenspointner 2005 ) ) .
the annihilation line signal is strong enough to allow more detailed studies of its galactic distribution .
@xcite fitted a variety of galactic bulge , halo , and disk models to the data .
stellar bulge and halo models describe the annihilation line emission from the central region of our galaxy equally well
. on adding a galactic disk component to the model fits , the faint annihilation radiation from the galactic disk is detected at the 3 - 4@xmath16 level . from these model
fits a total 511 kev line flux from our galaxy of @xmath24 ph @xmath18 s@xmath12 is derived , with the bulge - to - disk flux ratio being about 13 and the bulge - to - disk annihilation luminosity ratio being even larger with a range of 39 .
consequently , the model fits confirm the qualitative mapping result : spi observations demonstrate for the first time that the annihilation radiation from our galaxy is dominated by the bulge region ( ( * ? ? ?
* kndlseder 2005 ) ) . assuming a ps fraction of 0.93 the total 511 kev line flux corresponds to a galactic positron annihilation rate of @xmath25 s@xmath12 .
another approach to identifying the sources of galactic positrons is to compare the sky distribution of the 511 kev line emission with all - sky intensity distributions observed at other wavelengths .
the relatively best agreement is found at wavelengths dominated by emission from members of old stellar populations , but none of the tracer maps provides an acceptable fit to the data .
the sky distribution of positron annihilation appears to be unique , as it is even more bulge dominated than potential old positron source populations such as type ia supernovae , novae , or low - mass x - ray binaries ( ( * ? ? ?
* kndlseder 2005 ) ) .
both @xcite and @xcite searched for evidence for contributions from point sources on top of the extended emission , but none could be found in the 511 kev annihilation line and the ps continuum emission .
similarly , @xcite did not find any evidence for point sources of 511 kev line emission in the gc region using observations by ibis .
the most detailed spectroscopy to date of the annihilation radiation from the gc region has been performed by @xcite .
the line is found to be composed of a narrow and a broad component , consistent respectively with the expected line width in a warm ism for ps formation by radiative recombination and ps formation in flight .
this result is consistent with an earlier analysis by @xcite .
when fitting the annihilation spectra that are expected for the five standard phases of the ism to the spi spectrum ( see fig .
[ jean_annspc_fig ] ) , it follows that about half of the positrons annihilate in each of the warm neutral and the warm ionized phases .
possible fitted contributions from annihilations in cold gas , molecular clouds , or hot gas are not significant so far .
the importance of annihilations in the warm phases is consistent with our current understanding of positron propagation and annihilation and the gas distribution in the galactic bulge ( ( * ? ? ?
* jean 2005 ) ) .
@xcite find a value for the ps fraction of @xmath26 , consistent with pre - integral results , and with the spi results obtained by @xcite and @xcite .
the prospects for further nucleosynthesis studies with integral are bright .
as the mission continues , we expect to learn much more about the origin of positrons in our galaxy from ever - improving mapping , spatially resolved spectroscopy , and dedicated observations of individual candidate sources .
we also expect to study the galactic distribution of @xmath0al in detail by extracting distance information from line shifts , to investigate the galactic distribution of @xmath1fe which combined with studies of @xmath0al will provide new insights into massive star nucleosynthesis as well as the galactic star formation and sn rates , and to clarify the explosion physics of ccsne by observing selected snrs like cas a. | since its launch in october 2002 , esa s integral observatory has enabled significant advances to be made in the study of galactic nucleosynthesis . in particular , the imaging ge spectrometer spi combines for the first time the diagnostic powers of high resolution gamma - ray line spectroscopy and moderate spatial resolution .
this review summarizes the major nucleosynthesis results obtained with integral so far .
positron annihilation in our galaxy is being studied in unprecented detail .
spi observations yield the first sky maps in both the 511 kev annihilation line and the positronium continuum emission , and the most accurate spectrum at 511 kev to date , thereby imposing new constraints on the source(s ) of galactic positrons which still remain(s ) unidentified .
for the first time , the imprint of galactic rotation on the centroid and shape of the 1809 kev gamma - ray line due to the decay of @xmath0al has been seen , confirming the galactic origin of this emission .
spi also provided the most accurate determination of the gamma - ray line flux due to the decay of @xmath1fe .
the combined results for @xmath0al and @xmath1fe have important implications for nucleosynthesis in massive stars , in particular wolf - rayet stars .
both ibis and spi are searching the galactic plane for young supernova remnants emitting the gamma - ray lines associated with radioactive @xmath2ti .
none have been found so far , which raises important questions concerning the production of @xmath2ti in supernovae , the galactic supernova rate , and the galaxy s chemical evolution . |
let us consider the hamiltonian system @xmath5 with @xmath6 , called the _
potential_. system describes the motion of a particle in the plane submitted to the force field @xmath7 .
it always admits the so - called _ hamiltonian _ @xmath8 as a rational first integral .
the potential @xmath3 is called _ ( rationally ) integrable _ if system admits another rational first integral @xmath9 , functionally independent on @xmath10 . intuitively , the integrability of @xmath3 is equivalent to the fact that can be solved in explicit terms .
integrability is a rare phenomenon and it is in general a difficult task to determine whether a given potential is integrable or not . for _ homogeneous potentials _ in @xmath11 , _ necessary _ conditions for integrability
were given by morales - ramis @xcite and by morales - ramis - sim @xcite .
building on these works , we design in this article an algorithm which takes as input a _
family _ of rational homogeneous potentials @xmath12 depending on parameters @xmath13 and which computes a set of constraints on the parameter values @xmath14 that are necessary for the integrability of @xmath15 .
these constraints turn out to be of polynomial nature in @xmath16 .
there are several difficulties in this parameterized setting .
the first one is that the integrability constraints provided by the morales - ramis theory on which our whole approach relies , are expressed in terms of quantities ( eigenvalues of hessian matrices at darboux points , see section [ sec : preliminaries ] ) which are not easily accessible .
we circumvent this basic difficulty by using an equation that relates the eigenvalues , but this brings a new technical complication since the equation is of diophantine type .
a third difficulty is that the number of darboux points itself may depend on the parameters , leading to _ singular _ cases .
we follow a classical approach , inspired mostly by ideas in @xcite .
our contribution to the topic is effective and algorithmic , as we provide a complete , proven and implemented algorithm for the problem of computing necessary integrability conditions for planar parametrized homogeneous potentials , with precise output specifications .
our algorithm uses classical tools in computer algebra , such as polynomial ideal elimination based on grbner bases techniques .
an important feature is the use of ( complex ) polar coordinates to represent homogeneous potentials by univariate rational functions with parameters @xmath17 .
this change of representation considerably simplifies the computations and the proofs .
for instance , in polar representation , _ singular _ cases are those with non - generic multiplicity of the roots / poles of @xmath18 .
they are treated by our algorithm , which builds a tree containing each possible singular case .
this approach is related with comprehensive grbner bases @xcite , which are avoided here thanks to some a priori knowledge about singular cases . in summary ,
our strategy for computing necessary integrability conditions for @xmath3 consists in 4 steps : _ ( i ) _
rewrite @xmath3 in polar coordinates ; _
( ii ) _ set up a diophantine equation whose solutions belong to the so - called _ morales - ramis table _ ( that contains all possible eigenvalues of the hessian of @xmath3 at darboux points of @xmath3 ) ; _ ( iii ) _ solve this diophantine equation ; _ ( iv ) _ rewrite the condition of having prescribed eigenvalues at darboux points as polynomial conditions on @xmath19 .
some prior works used a similar strategy , but it was unclear which cases were possible to tackle , in particular for singular ones .
the approach was not fully automatized and this explains that results were only available for special families of potentials , for instance polynomials of small degree ( 3 or 4 ) @xcite , as the number of singular cases grows very fast ( already @xmath20 for polynomials of degree @xmath21 ) . by contrast , our treatment is unified and fully automated , and it allows not only to retrieve ( and sometimes correct ) known results , but more importantly , to treat potentials of degrees previously unreached ( up to 9 ) . by applying our algorithm to polynomial potentials
, we found three new cases admissible for integrability at degree @xmath21 ( but still not proved to be integrable ) , and various new families for higher degrees .
an even more striking application of our algorithm is the first complete proof of the non - integrability of the _ collinear three body problem _ , on which only partial results were known @xcite .
the direct approach that consists in searching first integrals @xcite is complementary to our ( non-)integrability analysis , as our algorithm helps either proving that the lists in @xcite are complete , or finding new unknown cases . .
( this is because the morales - ramis theory is much less powerful when @xmath22 . ) _ convention of notation : to avoid confusion , we will use bold letters for variables / parameters , and italic letters for parameter values_.
there exist strong integrability constraints ( see theorem [ thm : morales ] below ) .
they require to deal with darboux points , whose definition we now recall .
note that , by homogeneity , we could have chosen an arbitrary normalization non - zero constant on the right - hand side of . in the literature , this normalization constant is frequently chosen equal to @xmath26 @xcite .
however , our choice is deliberate , see the remark after theorem [ thm : morales ] . the following result ( which is an application of a more general criterion due to morales and ramis @xcite ) provides _ necessary _ conditions for integrability under the form of constraints on eigenvalues of hessian matrices at each darboux point .
it is the basic ingredient for numerous non - integrability proofs @xcite .
roughly , its main idea is as follows .
a darboux point leads to a straight line orbit of the dynamical system associated to @xmath3 , around which the system can be linearized .
if the whole system is integrable , then the linearized system , which in our case corresponds to a hypergeometric equation , is also integrable .
thus the integrability table of theorem [ thm : morales ] below is reminiscent of kimura s classification @xcite of solvable hypergeometric equations .
[ thm : morales ] ( morales - ramis @xcite ) let @xmath6 be a homogeneous rational function of homogeneity degree @xmath1 , and let @xmath27 be a darboux point of @xmath3 . if the potential @xmath3 is integrable , then for any eigenvalue @xmath28 of the hessian matrix of @xmath3 at @xmath29 , the pair @xmath30 belongs to the following table , for some @xmath31 . for polynomials of degree @xmath32 and @xmath33 ,
we retrieve known results ( leading to a complete classification of integrable homogeneous potentials of degree @xmath32 , and almost complete for degree @xmath33 ) . for @xmath34
, we obtain after simplification the following ideals @xmath35,[\a_4,\a_5],[36\,\a_5\a_1-\a_3 ^ 2,6\a_4\a_1-\a_3\a_2,6\a_2\a_5-\a_4\a_3],[44979\a_2 ^ 2-\\ 376712\a_3\a_1,66879684\a_5\a_1 - 75625\a_3 ^ 2,16719921\,\a_{{4}}\a_{{2}}-4708900\,{\a_{{3}}}^{2},\\ -376712\,\a_3\a_5 + 44979\,{\a_{{4}}}^{2},8178\,\a_{{4}}\a_{{1}}-275\,\a_{{3}}\a_{{2}},8178\,\a_{{2}}\a_{{5}}-275\,\a_{{4}}\a_{{3 } } ] , \\ [ -392\,\a_{{3}}\a_{{1}}+99\,{\a_{{2}}}^{2},484\,\a_{{5}}\a_{{1}}-{\a_{{3}}}^{2},1089\ , \a_{{4}}\a_{{2}}-196\,{\a_{{3}}}^{2},\\ -392\,\a_{{3}}\a_{{5}}+99\,{\a_{{4}}}^{2},22\,\a_{{4}}\a_{{1}}-\a_{{3}}\a_{{2}},22\,\a_{{2}}\a_{{5}}-\a_{{4}}\a_{{3 } } ] , \\ [ -40\,\a_{{3}}\a_{{1}}+7\,{\a_{{2}}}^{2},15876\,\a_{{5}}\a_{{1}}-25\,{\a_{{3}}}^{2},441\,\a_{{4}}\a_{{2}}-100\,{\a_{{3}}}^{2},\\ -40\a_3\a_5 + 7\a_4 ^ 2,126\,\a_{{4}}\a_{{1}}-5\,\a_{{3}}\a_{{2}},126\,\a_{{2}}\a_{{5}}-5\,\a_{{4}}\a_{{3 } } ] \end{split}\ ] ] the first two cases are the exceptional ones with @xmath36 , and the other cases lead indeed to integrable potentials .
these are exactly the conditions found in @xcite . at degree
@xmath21 , _ two non - trivial new potentials ( up to conjugation and rotation - dilatation ) are detected , not known to be integrable _ , but satisfying all integrability conditions .
their eigenvalue sets are @xmath37 .
the second has algebraic coefficients of degree @xmath38 , and the first is @xmath39 at degree @xmath40 and @xmath41 , the only possible cases are either already known , or @xmath36 , or they do not have darboux points .
thus for degree @xmath41 , the only cases whose integrability status is still unknown are up to rotation - dilatation @xmath42 and @xmath43 .
so we obtained a classification of integrable homogeneous polynomial potentials of degree @xmath41 . for degree @xmath44 and @xmath4 ,
some optimizations are necessary for the algorithm to be workable .
indeed , thanks to the fact that our family is invariant by rotation - dilatation , it is only necessary to consider functions @xmath45 with the coefficient @xmath46 and with the trailing coefficient of one polynomial factor equal to @xmath26 .
this removes two variables in the elimination ideal , and reduces by @xmath47 the hilbert dimension of the output . at degree
@xmath4 , we find _ three new cases satisfying all integrability conditions _ ; they are given by @xmath48 .
m. i. vigo - aguiar , m. e. sansaturio , and j. m. ferrndiz .
integrability of hamiltonians with polynomial potentials .
, 158(1):213224 , 2003 .
selected papers from the conference on computational and mathematical methods for science and engineering ( alicante , 2002 ) . | let @xmath0 be a rationally parametrized planar homogeneous potential of homogeneity degree @xmath1 .
we design an algorithm that computes polynomial _ necessary _ conditions on the parameters @xmath2 such that the dynamical system associated to the potential @xmath3 is integrable .
these conditions originate from those of the morales - ramis - sim integrability criterion near all darboux points .
the implementation of the algorithm allows to treat applications that were out of reach before , for instance concerning the non - integrability of polynomial potentials up to degree @xmath4 .
another striking application is the first complete proof of the non - integrability of the _ collinear three body problem_. * categories and subject descriptors : * + i.1.2 [ * computing methodologies * ] : symbolic and algebraic manipulations _ algebraic algorithms _ + * general terms : * algorithms , theory . + * keywords : * integrability , potentials , algorithms . |
in 2005 , three years after the first h.e.s.s . 12 m telescope was commissioned and one year after the fourth was installed , the decision was taken by french and german agencies to build a fifth instrument .
this new imaging atmospheric cherenkov telescope was mainly designed to obtain an energy threshold of @xmath1 , increasing the sensitivity of the array towards low energies @xcite . in order to achieve this goal , a new fast
, high sensitivity and low dead - time 2048 pixel camera was designed to equip the focal plane of a parabolic tessellated mirror of @xmath0 . with this new telescope , the h.e.s.s .
project was entering its second phase .
the goal of the h.e.s.s .
telescopes is to record and analyze the short and faint cherenkov light flash created by very high energy gamma - rays when they interact with the atmosphere and produce an extensive electromagnetic shower @xcite .
this technique , used by several experiments around the world , has allowed the detection and characterization of more than a hundred galactic and extragalactic objects so far @xcite . most of the high energy gamma - ray sources studied by h.e.s.s .
, as well as the hadronic background , have spectra which can be parameterized with a power - law in the range between tens of gev and tev energies .
this implies that lowering the energy threshold from hundreds of gev to tens of gev requires the data acquisition chain to be able to handle much higher trigger rates .
the dead - time has also to be reduced .
the camera of the fifth h.e.s.s .
telescope ( simply called `` ct5 '' in the following ) is a complete new design allowing to meet these challenging constraints : the dead time was lowered to @xmath2 and the trigger rate . ]
can reach @xmath3 . since it was designed several years after the first four h.e.s.s .
cameras ( in the following , the first four telescopes of h.e.s.s . will be referred to as `` ct14 '' ) @xcite , the new camera benefits from the latest progress in electronics integration : fpgas and fifo buffers are used extensively and dedicated asics were specifically designed for ct5 .
the paper is divided in two parts .
this first part gives a complete description of the ct5 camera .
the main improvements as compared to the ct14 technology will be pointed out when relevant .
the second part of the paper , to appear later , will deal with the performance of the camera in its nominal working configuration on - site in namibia .
the present part is organized as follows . in [ sec : conespm ] , the light guides , photomultiplier tubes and very front - end electronics are described .
[ sec : elec ] deals with the electronics , giving details on the front - end and the trigger as well as on the safety , slow control and monitoring ( [ sec : slc ] ) .
the mechanics of the camera are discussed in [ sec : meca ] , the calibration instrumentation in [ sec : calib ] and embedded software in [ sec : soft ] .
comes from the new coating used for ct5 cones.,width=336 ] the photomultiplier tube ( pmt ) which was chosen for ct5 is photonis xp-29600 .
this pmt is a minor update of the xp-2960 used for ct14 : the length of the pmt was reduced to decrease the ringing of the anode signal and the average quantum efficiency was improved from 25% to 30% .
the main characteristics of these pmts are cited in table [ tab : h1h2comp ] .
each pmt is soldered on a circular base where the resistive divider is located .
this base itself is connected to a board where the high voltage ( hv ) is obtained using a cockcroft - walton generator .
the high voltage can be set between 0 and @xmath4 with a @xmath5 resolution . to protect
the pmt from high light levels ( bright stars , shooting stars ) , the current ( noted hvi in the following ) drawn by the pmt is monitored and the high voltage is turned off by the board itself when the current exceeds @xmath6 , independently of the slow control and safety management system .
the circular base and the hv generation board have been designed and produced by the iseg company .
2500 pmts were tested for use in the ct5 camera , including spares .
the tests included gain calibration , measurement of the linearity and of after - pulse rate , as well as photo - cathode uniformity .
each pmt is equipped with a `` winston cone '' light guide @xcite to minimize dead - space between the pmt sensitive areas and to cut down on albedo light ( not coming from the mirror ) .
the use of winston cones in vhe gamma - ray imaging astronomy was introduced in 1994 @xcite , and is currently used by all imaging cherenkov cameras .
each winston cone guides photons which impinge on it with an angle below its `` cut - off '' angle ( defined by its geometry ) towards the pmt entry window after on average a single reflection , while reflecting back the photons above this cut - off angle . for a cut - off angle set to be close to the arrival direction of photons coming from the edge of the mirror ,
this provides almost complete protection from albedo light from the night - sky or from light sources and diffusive reflections from the ground . for ct5 ,
the cut - off angle used for the winston cones is about @xmath7 .
the same physical size of pmt is used as in the first cameras , which combined with the similar @xmath8 , allows an identical winston cone design to be used , for which details can be found in @xcite .
this incidentally gives a similar night - sky background ( nsb ) rate per pixel in ct5 as in the first four telescopes ( on average @xmath9100mhz of nsb photo - electrons ) , so the pmts can operate at the same gain .
the cones are assembled from injection moulded polycarbonate half - cones with aluminization and mgf@xmath10 protection , and have a hexagonal symmetry around their axis , to minimize the dead - space between the cones .
the winston cone entry aperture ( at the mirror - facing side ) of @xmath11 flat - to - flat on each hexagon defines the pixel field of view ( fov ) of @xmath12 ( compared to @xmath13 for the first cameras ) .
the wall thickness of the cone entry is @xmath14 , to minimize the dead area in the focal plane . in order to reduce their cost and to improve their performance reproducibility ,
they were produced by an industrial process optimised for this production at the company . in order to validate the light guide design and to measure each cone after the mass production ,
the automatic test bench used for the winston cones of the first cameras was re - used .
measurements of the cut - off angle and of the absolute optical transmission are made with a diffusive source covering the same angular aperture as the ct5 mirror as seen from the camera .
2669 cones have been measured following the production .
[ fig : lctrans ] shows their average absolute optical transmission as a function of the wavelength , defined as the ratio of the amount of light detected by a photo - sensor to that entering the winston cones .
while the transmission is about the same for ct14 and ct5 cones above @xmath15 , it has been improved by 15% below that wavelength in the range detectable by the pmt .
this difference is due to the new coating used for ct5 winston cones and should result in an improvement of the global optical efficiency of the telescope by a few percent . as the winston cone geometry is strictly identical that used in ct14 , the measured transmission for ct5 as a function of the angle of incidence is identical to that of ct14 ( see figure 13 of @xcite )
the front - end of the camera is modular , made - up of 128 identical electronics modules ( hereafter called _ drawers _ , fig .
[ fig : photodrawer ] ) inserted in the camera from the front so that they can easily be replaced .
each drawer carries 16 pmts ( for a total of 2048 pixels ) and three electronics boards : two boards for data acquisition ( level 0 trigger ( l0 ) , amplification , sampling , conversion of pmt signals ) and one board for slow control .
the drawers can receive and send data through dedicated communication buses , using a token - passing mechanism .
one bus ( parallel , @xmath16 ) is devoted to data transfer and the other to slow control messages ( serial , @xmath17 ) .
there are 16 data buses of 8 drawers each , and 8 slow control buses of 16 drawers each . at the rear of the camera , three compact pci ( cpci ) crates
are used for data management , trigger , safety and slow control .
each of these tasks is managed by a cpu , except for data management for which two cpus are used to optimize data transfer and sustain the required trigger rate of @xmath3 .
other equipment include five power supplies , a gps receiver used for time stamping , a terminal server used to access the consoles of the cpus for maintenance as well as the gps serial interface .
all the cpus , the power supplies and the terminal server are connected to each other and to the computers on the ground through a gigabit ethernet switch .
while normally located in the focal plane during observations , the camera can be unloaded from the telescope and moved inside a shelter during full - moon period or for maintenance and calibration operations . a dedicated
, fully - automatic loading / unloading system was designed for this purpose . as a consequence
, the number of connections going from the camera to the ground has been minimized to a single @xmath18 power cable , 3 optical fibres for data transfer and trigger , and a compressed air tube .
electrical and optical links are grouped in a single connector .
the front - end electronics consists of three boards in each drawer ( figure [ fig : drawer ] ) .
two boards ( analogue memory boards , `` analo '' ) are used to process the signals from the pmts ( 8 pmts per board ) and one board ( slow control board , `` slow '' ) is used for the control and monitoring of the two analo boards connected to it . the sam chip is a two - channel analogue memory , based on switched capacitors , that is used as a buffer , in which the photomultiplier tube signal is sampled and stored while waiting for a decision from the fixed - latency trigger electronics .
this chip , described in more detail in @xcite , has been specially designed for ct5 to overcome the main limitations of the ars chip @xcite equipping the cameras of the first h.e.s.s .
sam can operate with sampling frequencies in the range from 0.5 to @xmath19 ( giga - samples per second ) with a dynamic range exceeding 11 bits .
this high signal over noise ratio is obtained thanks to a full differential architecture rejecting the pick - up noise coming from the digital activity inside the chip and on the front - end boards .
the chip memory depth has been extended from 128 to 256 cells to be compatible with trigger latencies higher than @xmath20 for a @xmath21 sampling frequency .
the chip analogue bandwidth , increased from @xmath22 to @xmath23 , is fast enough to avoid the widening of the fast pulses from the photodetectors .
it allows for use of short integration windows to calculate the pulse s charge , minimizing the effect of the night sky background ( nsb ) .
an spi slow - control link permits to program many parameters and to access to the various modes of operation of the chip . for each event , only few samples , corresponding to a region of interest within a window located at a fixed time ( programmable using a slow - control parameter ) with respect to the trigger , are read out .
the consecutive samples from the two channels are multiplexed in parallel toward an external 10 msps 12-bit 2-channel external adc for digitization .
for a 16-sample event , the readout time is only @xmath24 , instead of @xmath25 with the ars chip .
this major improvement allows the camera dead time to be decreased below 1% for a typical trigger rate of @xmath3 .
because of its very good and stable cell - to - cell uniformity of pedestal and gain , the sam chip only requires a very simple calibration .
this operation consists of equalizing 32 dc values using on - chip spi digital to analogue converters to cancel offsets due to the matrix chip structure .
once this operation has been performed , it remains valid ( less than 10% change of noise ) for a few days and for temperature variations within a range of @xmath26 , with the digitized sam output data being directly usable for online processing by the fpga of the analogue memory board , without additional correction .
a batch of 6000 sam chips has been produced using a @xmath27 cmos process from ams .
each chip has been tested using a fully - automated procedure to ensure its functionality and compatibility with ct5 requirements .
this procedure includes power consumption , baseline and noise measurements but also tests performed using input pulses simulating the photomultiplier response such as signal shape , linearity , jitter and crosstalk characterizations .
the on - chip time delay between the trigger and the signal paths defining the timing precision for the beginning of the region - of - interest was also extracted .
the uniformity of this parameter is better than @xmath28 over the whole batch .
all the parameters extracted during the tests were recorded in a database for further chip sorting and analysis . to guarantee its stability ,
the test bench was monitored over the 6 months of tests by retesting a set of reference chips every day .
fewer than 10% of the chips were found to be out of specifications .
the negative pulse from a pmt is amplified ( or attenuated ) with three different gains : a gain @xmath29 ( high gain ) used for low energy events ( typically under 400 p.e . ) and to measure the single p.e .
response for pmt calibration , a gain @xmath30 ( low gain ) used for high energy events ( up to 5000 p.e . ) and a gain @xmath31 , used for the trigger signal generation . in addition , the signal is inverted in high gain and low gain channels .
high - gain and low - gain signals ( fig .
[ fig : display]a , [ fig : display]b ) are stored into two 256-cell analogue memories ( within the sam chip , [ subsubsec : sam ] ) , acting as circular buffers .
each cell corresponds to a @xmath28 sample of the signal .
if no trigger occurs , the analogue memory content is overwritten as new data are recorded .
if a trigger occurs , the buffering is stopped and a given number of cells in the memory is read and digitized .
this number is programmable and defines the width of the readout window , which depends only on the pmt pulse width and nsb .
for all h.e.s.s .
cameras , it is set to 16 cells , corresponding to @xmath32 .
the digitization is done by a 12 bit , @xmath33 adc and the output digital is in the range @xmath342048 ( decimal ) . ] . in order to use the full dynamic range of the adc , the digitized signal ( positive pulse )
is shifted towards negative values , fig .
[ fig : display]a and fig .
[ fig : display]b . ] .
the 12 bits , corresponding to one @xmath28 sample of the pmt signal , are then sent to a @xmath35 first - in first - out ( fifo ) chip .
this buffer allows to store new events ( to a maximum of 64 ) while waiting for a l2 decision signal ( _ l2 accept _ or _
l2 reject _ , [ subsec : trigger ] ) and is therefore extremely important to reduce the overall dead - time .
each analo board has a fpga chip performing several operations .
first , it is used to control the sam chips . in particular
, it stops the buffering and starts the digitization when a l1 trigger occurs .
after digitization , the data are stored in the fifo buffers .
second , the fpga also controls the readout of the fifos and processes their content when a l2 trigger happens . from the 16 signals provided by the pmts ( 8 pmts @xmath36 2 gains )
, the fpga builds the data blocks in the mode specified during the configuration of the camera .
two modes can be used : the `` sample mode '' , for which the fpga sends all the samples for each pmt for the two gains , and the `` charge mode '' , for which the fpga integrates the pulse and sends the total charge , as an adc count coded on 16 bits , for the two gains .
in addition , the fpga computes the time of the maximum of the pulse ( @xmath37 , fig .
[ fig : display]c ) with respect to the l1a trigger time and the time spent by the signal over a given absolute threshold ( tot , for _ time over threshold _ , fig .
[ fig : display]d ) .
[ fig : spe ] shows the charge distribution obtained through the high gain channel when a pmt receives an amount of light which results in the occasional emission of one single photo - electron ( spe ) .
following @xcite and according to what was done previously in h.e.s.s .
@xcite , the distribution is fitted with a sum of gaussian curves with poisson - distributed relative weights to extract the mean of the gaussian corresponding to the spe relative to the position of the pedestal ( noted @xmath38 ) , the standard deviation of the pedestal ( @xmath39 ) and the standard deviation of the spe ( @xmath40 ) .
the pedestal variation that can be seen in the figure is mainly due to the pmt and the amplifier and has a standard deviation of 10 adc counts ( @xmath41 ) .
the sam and the adc contribute only marginally with a noise standard deviation of @xmath42 and @xmath43 respectively . and
@xmath40 ) , and the separation between the two peaks ( @xmath38).,width=336 ] the slow control board ( slow ) is used to control all the parameters of the drawer and to encapsulate these parameters for their transmission in the data pipeline .
some quantities are obtained directly by the slow board ( e.g. , the temperature , hv and hvi . ] for each pixel ) and some others through the fpga located on each analo board ( e.g. , the l0 trigger threshold , the trigger rate for each pixel , @xmath37 and tot values for each pixel ) .
as in the first four cameras , each pixel can provide a level 0 ( l0 ) trigger and the readout of the whole camera is triggered by a level 1 ( l1 ) decision if several pixels are triggered in a contiguous sector of the focal plane .
for the fifth camera , an additional level of triggering was introduced , a level 2 ( or l2 ) topological trigger .
the l2 trigger is used to reduce the data rate and select gamma - like events . during regular data taking
, the camera uses the l1 trigger with l2 activated or not , while during calibration runs such as single photo - electron runs , the trigger can be provided by an external source , or generated by the software .
given the fact the fifth telescope has a lower threshold and a higher event rate than the others , it triggers alone most of the time75% of the events . ] .
the cameras of the five telescopes are connected to the central trigger of the array @xcite through optical fibres .
the central trigger hardware , located in the control - room building , receives trigger signals from the five cameras and provides an event number to the cameras triggered in coincidence . in the case of ct5 , the central trigger also informs the camera if it has triggered alone ( monoscopic event ) or in coincidence with another telescope ( stereoscopic event ) .
this information is used by the l2 trigger .
the l0 trigger is produced on each analo board ( fig .
[ fig : trigger ] ) . the trigger signal for each pixel
is compared to a given threshold @xmath44 by a comparator .
the threshold ( expressed in p.e . ) is programmable and the same value is set for all the pixels .
the outputs of the eight comparators of a analo board ( one per pixel ) are summed and sent to dedicated boards in the trigger crate .
it is possible to enable or disable a pixel so that it participates in the trigger decision or not .
a pixel is disabled when the hvi exceeds @xmath45 .
this value is reached typically when a star of magnitude 78 enters its field of view .
the value of @xmath44 is set so as to minimize the influence of the nsb on the trigger rate . to determine @xmath44 ,
the pmts are illuminated with a continuous white light , which represents the nsb . for h.e.s.s .
, the nsb rate seen by one pixel can vary from @xmath46 to @xmath47 ( p.e . ) depending on the region of the sky observed , for an average of @xmath48 .
[ fig : afterpulse ] shows the evolution of the l0 trigger rate as a function of the l0 threshold .
two contributions can be seen : below 5 p.e .
the nsb dominates , and above this charge the pmt after - pulses dominate .
the l0 threshold is set just below the break of the curve .
the present setting is @xmath49 . for the next - level trigger ( l1 )
, the drawers are grouped in sectors ( fig .
[ fig : sectors ] , fig .
[ fig : display]e ) .
a sector is a group of 64 contiguous pmts , with a vertical and horizontal overlap of 16 and 32 pixels respectively .
the l1 trigger is obtained when a sector has at least @xmath50 pixels satisfying the l0 condition .
@xmath50 is currently set to 3.5 .
the trigger decision is issued by a signal called l1a ( _ l1 accept _ ) in the following . when a l1a occurs , the data stored in sam memories are read out , digitized , and buffered in the fifos . at this point ,
the sam chips are available for a new event .
therefore , the use of fifo buffers in the front - end pipeline allows to release the l1 trigger just 7 to @xmath51 after the l1a .
this is a major improvement as compared to the value of @xmath52 measured on the ct14 cameras , for which no buffering is used . during
the readout of sams , other triggers can happen , but the camera is not read - out and a `` busy '' state is transmitted to the central trigger .
this information is used to calculate the dead - time of the whole array .
the parameters @xmath44 and @xmath50 are stored in a database and loaded in drawer fpgas when the camera is configured at the beginning of each night .
different values can be used depending on nsb conditions .
as already mentioned , monoscopic events ( i.e. , events triggered by ct5 only ) constitute a large fraction of the l1a rate of the ct5 telescope . among the latter ,
only very few are gamma events , so the purpose of the l2 trigger @xcite is to reject as much as possible of the residual nsb and of the hadronic background .
the 2048 camera pixel signals go through the l0 comparators and a second comparator with threshold @xmath53 .
the comparator outputs are shaped and stored in a delay - line buffer implemented within a dedicated pre - l2 asic in order to compensate the l1 trigger latency .
when a l1a occurs , a ternary image giving the list of pixels above @xmath54 and @xmath44 , resynchronized with respect to the l1a signal , is transmitted to the l2 trigger system ( fig .
[ fig : display]f ) .
the l2 behaviour depends on the type of the event : monoscopic or stereoscopic .
this information is provided by the central trigger . in the case of monoscopic events ,
the ternary pixel map transmitted to the l2 trigger is processed by a real - time implementation of the l2 selection algorithm based on the detection of clusters of active pixels , and the computation of first and second order hillas moments @xcite to identify gamma - like or hadron - like images . in the first case ,
a signal l2a ( _ l2 accept _ ) is generated , which triggers the read - out of the fifos for the whole camera .
otherwise , an l2r signal ( _ l2 reject _ ) is issued to reject the current event : the event is dumped from the fifos and discarded , and the l2 system is ready to process a new l1a event . in the case of stereoscopic events
, the l2 trigger skips the selection algorithm and generates a l2a signal . in other words ,
a l2a is generated for all events triggering more than one telescope .
when a l2a occurs for an event , all the data corresponding to that event are read from the drawer fifos , for all pixels .
this operation is performed by the fpgas and requires @xmath55 in total .
the readout is done in parallel for the sixteen data buses .
then , all data are sent to fifos located in the data management crate .
this second level of buffers is used to store the data blocks until they are merged to form the full event and sent to a farm of computers located in the control - room building , for further processing @xcite .
the overall dead - time of @xmath2 quoted in table [ tab : h1h2comp ] includes the time to release the trigger plus the time interval necessary for the central trigger to send the event number and event type ( monoscopic or stereoscopic ) .
the safety and slow control system monitors the camera environment parameters ( temperature , level of ambient light ) and the status of different camera sub - devices such as the position of the lids , the position of the mylar plate ( see [ sec : meca ] and [ sec : calib ] ) , etc .
it also controls the fans and allows to enable or disable the opening / closing of the lids , the pmt high voltage and the camera power supplies .
the outputs of the positioning leds ( [ sec : calib ] ) , seen by a ccd camera located on the dish , can be controlled individually .
these outputs are used to measure the camera position precisely .
in addition , the system includes safety functions that protect the camera in case a failure in critical components is detected . for this purpose , the system is able to take fast , standalone decisions . for example , the system can make an emergency stop of the pmt power supply , or prohibit the opening of the camera lids if the safety conditions are not fulfilled .
this main logic is implemented in a fpga hosted in a cpci crate placed at the rear of the camera . due to the high number of required interfaces and connections ,
the safety and slow control system is implemented in several boards .
one board is dedicated to the main logic , the cpci interface with the cpu crate and the photo - diode reading .
a second board drives the positioning leds .
three others are dedicated to the fan control and the last one to the camera temperature monitoring . even though the safety and slow control hardware logic is located inside the camera and interfaces directly with the different devices , the system is piloted remotely from the h.e.s.s .
array data acquisition system @xcite through a cpu board .
thus , the slow control can be configured with different levels of safety .
in addition , all monitored data or the standalone decisions taken locally can be reported .
as soon as the camera is powered , the safety system is active .
the mechanics of the ct5 camera relies on the same general concepts successfully applied in the previous cameras . because of the much larger size of this camera relative to those of ct14 , slight adjustments of the camera concept have been made and the experience gained in building the ct14 cameras allowed us to improve the mechanical design in some respects .
the requirements of the ct5 camera are given in table [ tab : h1h2comp ] , with a special care on the minimization of the shadowing by the camera , improved reliability of the electronics cooling and improved accuracy of the pixel location .
the camera is attached to the telescope quadrupod such that its distance to the dish can be remotely controlled to adjust the image plane to focus on the shower maximum ( [ subsec : autofocus ] ) .
it can also be unloaded and loaded remotely by a trolley system to allow it to be parked into a shelter for maintenance and protection . as a consequence , the camera does not need to be fully waterproof .
the fully equipped camera has a total weight of 3 tonnes , which corresponds to the weight budget allowed by the design of the mechanical structure of the telescope .
and its outside diameter with a partial coverage is @xmath56 . for comparison ,
the lighter grey area in the centre corresponds to the ct14 camera layout ( same physical scale).,width=336 ] the heart of the detection system of the camera is located close to the focal plane ( fig .
[ fig : camerafront ] ) and hosts the photo - detection system composed of 2048 pixels arranged as shown in fig .
[ fig : fp ] and consisting of the winston cones , the pmts ( [ sec : conespm ] ) , and their associated electronics , i.e. , the front - end boards ( [ subsec : fe ] ) . the pmts and the front - end boards are embedded into a modular mechanics structure called a _ drawer _
[ fig : photodrawer ] ) .
they can be manually extracted from the camera front , which requires to unscrew just two screws .
this system , very similar to the drawer concept used for ct14 , brings a high modularity for the assembly , the integration , the use of interchangeable drawers and the maintainability . the power and electronics connections with the rest of the camera
are made via three connectors placed on the rear of the drawers .
these connections impose that the drawers must be positioned with an accuracy better than @xmath57 . to form the focal plane and to insure a good electrical connection when the drawers are inserted from the front
, the drawers are placed into the central part of the mechanics , called the `` sandwich '' ( fig .
[ fig : sand ] ) .
it is made of two thin aluminium sheets ( @xmath58 ) assembled from 146 aluminium profiles ( with a length of @xmath59 ) , developed specifically for this camera . with this system
, the drawers location within the focal plane has an accuracy of @xmath60 . by design ,
this sandwich s second function is to provide mechanical stiffness to the whole camera .
the management of all the interfaces ( mechanical , electrical , optical ) was thus an important aspect during the design process .
the system of winston cones is placed in front of the sandwich and drawers .
they are held and placed within the telescope optical focal plane by a large aluminium plate ( @xmath61 ) , on which the winston cones are clipped with plastic washers . as the cone localisation determines the astrometric accuracy of the pixels and the optical transmission between them and the pmts , this aluminium plate is machined with a high accuracy such that the winston cones are localised at @xmath62 on it ( or 0.01 arcsec ) .
its distance to the drawers ( @xmath63 ) is finely controlled by screws distributed over the whole surface of the plate and limiting the effect of its gravity deformation .
these accuracy values have been measured by a company specialised in metrology .
the instrumentation of the focal plane and all the camera equipment and services are integrated into the camera body ( fig .
[ fig : explodedview ] ) .
this is a skeleton of thin aluminium profiles , whose shape has been specially designed for the project .
the profiles are welded together to avoid any ageing .
a company specialised in producing and integrating large and complex aluminium pieces assembled these profiles to the required accuracy , reaching a measured @xmath64 of the external dimension of the camera body .
the camera envelope , made with thin aluminium plates , is riveted on this skeleton .
the plates provide a thermal isolation from the direct sun - light power .
this envelope integrates also a system for cooling , as described in [ subsec : cooling ] .
this skeleton holds also all the internal equipment , the sandwich with its 128 drawers , the two 19-inch electronics racks carrying electronics boards and services such as the power supplies , and the front , lateral and rear lids .
it should be noted that front lids have several functions .
first , they protect the focal plane from the outside environment and second , they hold calibration instrumentation ( [ sec : calib ] ) .
they are motorized by pneumatic pistons and controlled either manually or remotely by the camera slow control system ( [ sec : slc ] ) .
the pressurized air is generated by an on - ground air compressor and is brought to the camera by pneumatic tubes fixed on the telescope structure .
three air - tanks are used along the path to create buffer volumes for safety .
the choice of a pressurized air system fully managed by pneumatic logic has been driven by the need to be able to close the camera lids automatically in case of a general power failure .
finally , the skeleton supports the two large mechanics parts making the interface to the telescope quadrupod structure ( [ subsec : autofocus ] ) . during the design process ,
finite element simulations ( fem using the ansys module within the catia software @xcite ) have been carried out by modelling mechanics pieces as hollow components .
this allowed to optimise the design with a focus on the ratio of stiffness to weight and to achieve a rigid and accurate body .
this process led us to choose components made of aluminium profiles welded together , despite the fact that only a few companies have the technology and the know - how to produce such large objects .
the main camera dimensions are summarized in table [ tab : h1h2comp ] .
the electronics components dissipate about @xmath65 of thermal power inside the camera . for proper operation and extended lifetime of the electronics ,
this heat must be evacuated so that the temperature of the camera is kept as stable as possible during data taking . as for the ct14 cameras , the cooling is obtained by convection and by forced - air flows .
fresh filtered air is injected inside the camera by large fans located on the camera envelope and protected from rain by lightweight hoods .
this fresh air is convected to the rear of the sandwich . for each drawer ,
two fans blow the air towards the electronic boards .
then the warm air is drawn by large fans that blow it out of the camera .
the speed of the different groups of fans can be changed under slow control , depending on atmospheric conditions .
this ventilation system , despite its simplicity , allows to efficiently cool the fe electronics and the instrumentation located into the racks at the rear of the camera .
the nominal distance between the position of the camera and the centre of the mirror is @xmath66 , which corresponds to the focal length of the telescope .
however , as the atmospheric showers develop at a finite altitude between @xmath67 and @xmath68 , the position of the camera can be adjusted to optimize the trigger rate and angular resolution depending on observation conditions ( in particular the zenith angle ) @xcite .
the change in position is done thanks to a system which can move the camera along the optical axis of the telescope .
this focusing system has been designed to operate in the sometimes harsh environmental conditions of the namibian site .
its main technical specifications are given in table [ tablefocus ] .
.[tablefocus ] main requirements of the focusing system . [ cols=">,^ " , ] the mechanical design of the focus system aims at two main functionalities : * install and lock the camera into the telescope quadrupod structure with an accuracy of @xmath69 . as mentioned earlier , the camera can be loaded or unloaded from the telescope for maintenance or calibration purposes .
the camera is locked and unlocked in the focal plane by using a pneumatic system composed of four toggle fasteners and four jacks ( fig .
[ fig : autofocaus ] ) . *
adjust the position of the camera along the optical axis , within the camera support mast structures .
the focusing system is made of two components , one is fixed to the telescope structure while the other can move in order to allow the translation of the camera .
this motion is driven by two ball screws and two brushless motors which guarantee a speed of @xmath70 and a position accuracy of @xmath71 .
the camera can be moved by @xmath72 .
the data acquisition software controls the camera position during the focusing process .
while the translation has started , this process has to be guaranteed until the end of the programmed movement , even in case of an unexpected stop ( e.g. , due to a power cut ) .
the system is controlled through a programmable logic controller architecture composed of two cpus , two field - buses and two specific variators used to react to power cuts and to manage synchronized operations of the two motors .
the first estimations , obtained through simulations , have shown that adjusting the focus height to the altitude of the shower maximum allows an increase of 510% of the trigger rate and a 10% improvement on the angular resolution at low energies ( @xmath73 ) , without any alteration of the performances at higher energies .
to be able to properly analyze and interpret the data produced by the camera , it is necessary to be able to monitor the gain of each channel , the uniformity of the pmt responses , and the exact position of the camera in the focal plane .
the conversion factor between adc channels and charges is monitored by illuminating the camera such that each pmt s photo - cathode emits a single p.e .
( spe ) , on average .
this calibration can be done either with the camera in the shelter or outside when the camera is in the focal plane . in the first case , a pulser
is used to generate flashes of blue light ( @xmath74 ) at the required intensity and rate , and the light illuminates the pmts directly through a diffuser .
the same pulser is used to trigger the camera via the external trigger input .
when the camera is in the focal plane , it is necessary to filter out the light from the nsb . for this ,
mylar sheets are placed in front of the winston cones , providing an attenuation of light intensity by a factor @xmath75 . as a consequence , to illuminate the pmts at a spe level , the intensity of the calibration light source , located at the centre of the telescope dish , has to be far greater than the intensity of the source in the shelter . the mylar sheets ( fig .
[ fig : camerafront ] ) are held by two large aluminium plates that can either be in contact with the cone plate during spe runs , or kept attached to the front lids by the use of small pneumatic clamps during normal observations .
when in the focal plane , the camera is triggered by an additional pmt , not covered by the mylar plate , located on the top of the focal plane .
when the flasher illuminates this calibration pmt , it delivers a signal that , shaped and delayed , is used as a trigger signal .
the calibration light source is used to provide either spe pulses or flat fielding patterns , used for pixel inter - calibration .
a @xmath76 pulsed laser ( @xmath77 ) is installed in a stainless steel container mounted at the centre of the telescope dish .
this container is under constant overpressure to avoid environmental pollution .
the repetition rate of the flashes is software selectable and set at @xmath78 by default .
a holographic filter expands the beam to a flat distribution over the camera fov .
a number of neutral density filters , mounted in front of the laser on a motorized filter wheel , are used to vary the intensity of the light output .
a dedicated processor board is mounted inside the camera housing and ensures control functions and communications with the daq .
the role of this board is to control the laser , the filter wheel , as well as the delay and the threshold applied to the calibration pmt signal to deliver a proper external trigger signal to the camera . for pixel inter - calibration ( the _ flat fielding _ )
, the camera is illuminated directly by the laser ( i.e. , the mylar plates are on the opened front lids ) at an intensity of @xmath9200 p.e .
the non - uniformity of the light intensity does not exceed 12% ( standard deviation ) . the charge seen by each pmt
is analyzed off - line and correction coefficients are applied to each pixel to compensate for collection and quantum efficiency inhomogeneities .
it is necessary to have the exact position of the camera in the focal plane in order to build an accurate model of the pointing and optical system for use during the data analysis @xcite .
this has a direct effect on the point spread function of the instrument .
16 red leds ( @xmath79 ) , with intensity controlled individually by the slow control system ( [ sec : slc ] ) , are accurately fixed on the winston cone plate , and are monitored during observations with a dedicated ccd camera located at the centre of the dish . in that case , the camera lids are opened .
when taking measurements for the pointing correction model and for the main mirror facets alignment , the camera front lids are closed and used as a screen to project the images of stars . in that case , the light from the leds passes through the lids via specific holes which , in other circumstances , are closed by movable covers remotely controlled by pneumatic actuators .
from the software viewpoint , the ct5 camera functionally consists of an array of 2048 photomultipliers which take the digital images , and which are serviced by various equipment such as power supplies , front - end electronics , triggering apparatus , gauges , fans , actuators , etc .
the control of the whole apparatus must be done remotely through a graphical user interface running on a control computer , and the different types of data generated during a run must be sent back to a central farm of analysis computers . the embedded software of the ct5 camera interfaces between the camera hardware and the array control room , to which it is connected via a gigabit ethernet link .
it addresses the following : programming and monitoring the various hardware parameters ( `` slow - control '' ) , acquisition of the shower image data , acquisition of the trigger data , and real - time on - board control of the camera security .
each of these points is handled by dedicated branches comprising a cpci crate , a processor board , miscellaneous proprietary interface boards and buses . as seen by the software ,
a branch is a server application running on the branch processor ( under a standard linux system with posix real - time extensions ) and servicing requests sent by the camera controller software as a client .
the server of the slow - control branch is particular because , besides being in charge of controlling the hardware parameters , it is used as a gateway for communications between the control computer and the three other servers .
each server is organized as a multi - thread process .
the scheduling policy of the different threads is set so as to ensure the deterministic behaviour needed for real time applications .
in such a scheduling policy , the running thread keeps running indefinitely until either it voluntarily relinquishes the control of the processor or until it is pre - empted by a thread of greater priority . in each server
, one thread listens to an ethernet socket in order to read the commands sent by the controller and a variable number of other threads execute the requested tasks .
the thread listening to ethernet is given the highest priority of execution .
the command and control server has three principal tasks to execute periodically ( in addition to several transitory on - request ones ) : read the gps time and distribute it to the data and trigger branch servers , so that they can remain correctly synchronized ; monitor the front - end electronics voltage , current and temperature , and shut down the power supplies if anything goes wrong ; ensure that the on / off status of each photo - multiplier is properly managed throughout the run , according both to the predicted movement of the stars through the detector and to any transient or accidental event that may occur during the programmed observation ( e.g. , shooting stars ) .
the server dedicated to image data acquisition has only one important task to do : read quickly the data from the drawers and send them as formatted blocks to the farm of analysis processors via a 1 gbits / s ethernet link . in order to avoid overloading the data receiver ,
several nodes are used and a switch of the receiving node is made every 4 seconds . to be able to sustain high acquisition rates ( more than 6 khz ) , the server is currently distributed over two separate processors operating alternatively , one reading the data while the other is writing them . this way , constant data reading can be maintained whatever happens on the receiver side .
this is a crucial feature to avoid the loss of data coherency that may happen if data overflow occurs at the front - end level .
synchronization of the two processes is carried out through a set of customized `` test - and - set '' devices implemented on the proprietary interface board .
the server can automatically reconfigure itself to execute on a single processor if necessary .
this is possible , however , only if the acquisition rate is relatively low ( less than 2 khz ) and if the network is sufficiently stable .
the server dedicated to the acquisition of the `` trigger '' data is similar in functionality to the data acquisition server but it is structured differently because it always has to run on a single processor .
besides the thread listening to the ethernet socket , it is composed of two threads instead of one .
one low - priority thread reads and formats the data from the trigger front - end boards ( hereafter called the `` reading thread '' ) and the other ( the `` sending thread '' ) sends them with higher priority , packed into multi - events bunches , to the farm of processors . as writing to the socket
is done in synchronous mode , it may happen that the sending thread gets stuck while waiting for the end of a data transfer . in that case
the reading thread simply goes on reading without trying any longer to transfer the data .
this allows to guarantee the events integrity , even when the trigger rate is very high .
besides its main function of data transfer , the server also executes miscellaneous on - request transitory tasks related to the management of the trigger set - up .
regarding calibration , the embedded software ( running on dedicated cards : one controlling the source unit and the other the trigger unit ) has also to implement the interfaces between the calibration hardware on the one hand and the daq processes on the other hand , in order to program and monitor the various hardware parameters .
the most important requirement of the ct5 camera is to be able to handle a trigger rate of several khz without any data corruption , i.e. , a factor of ten higher than the previous h.e.s.s . cameras .
this goal was achieved through a complete re - design of the electronics and of the acquisition pipeline . in particular ,
the dead - time was reduced significantly using new analogue memories , introducing fifo buffers at several stages of the acquisition chain and optimizing the data transfers .
the ct5 camera brings a lot of improvements , not only for the main acquisition electronics .
ct5 is the first cherenkov telescope in the world to actively move the focal plane in order to adjust to the depth of the shower development in the atmosphere .
the camera can be also unloaded from the telescope focal plane to protect it from the weather conditions or for maintenance operations .
many of these features could be considered seriously for a future project like the c.t.a .
array @xcite .
the ct5 camera was delivered on site in namibia in spring 2012 . since then , the new telescope was fully integrated in the h.e.s.s . array and data are being accumulated to fully understand the camera response , the efficiency of the trigger decisions and the acquisition system performance .
this testing phase consists of taking calibration runs ( flat field runs , pedestal runs , spe runs ) as well as regular observation runs on selected `` benchmark '' targets .
the results of this operating period will be described and discussed in the second part of this paper .
we would like to acknowledge the support of our host institutions .
we also want to thank the whole h.e.s.s .
collaboration for its support , and the referees , who helped us a lot improving the quality of the draft .
m. punch , proc .
`` towards a network of atmospheric cherenkov detectors vii '' , ed .
b. degrange , g. fontaine , ecole polytechnique , 2005 .
m. holler , et al .
33rd icrc ( rio de janeiro ) , 2013 .
t. weekes , aip conference proceedings 1085 ( 2008 ) 3 . j. a. hinton & w. hofmann , annual review of astronomy and astrophysics 47 ( 2009 ) 523 . http://tevcat.in2p3.fr m. punch , proc .
27th icrc ( hamburg ) , 2001 .
p. vincent , proc .
28th icrc ( tsukuba ) , 2003 .
f. aharonian et al .
( h.e.s.s .
collaboration ) , astroparticle physics 22 ( 2004 ) 109 .
r. winston & w. t. welford , `` high collection for nonimaging optics '' , academic press ( 1989 ) .
m. punch , proc .
`` towards a major atmospheric cerenkov detector iii '' , ed . t. kifune .
tokyo , japan , universal academy press , 1994 .
k. bernlhr et al . , astroparticle physics 20 ( 2003 ) 111 .
e. delagnes , et al . , nuclear instruments and methods a 567 ( 2006 ) 21 .
e. ballamy et al . , nuclear instruments and methods a 339 ( 1994 ) 468 .
a. wright , nuclear instruments and methods a 579 ( 2007 ) , 967 .
d. lachartre & f. feinstein , nuclear instruments and methods a 442 ( 2000 ) 99 .
s. funk , et al .
, astroparticle physics 22 ( 2004 ) 285 .
a. m. hillas , proc .
19th icrc ( la jolla ) , 1985
. y. moudden , et al . ,
astroparticle physics 34 ( 2011 ) 568 .
y. moudden , et al .
, ieee transactions on nuclear science 58 ( 2011 ) 1685 .
a. balzer , et al . ,
astroparticle physics 54 ( 2014 ) 67 .
r. mirzoyan , v. fomin & a. stepanian , nim a 373 ( 1996 ) 153 .
s. gillessen , for the h.e.s.s .
collaboration , proc .
cosmic ray conf . , tsukuba ( 2003 ) , univ . academy press , tokyo ,
the cta consortium , experimental astronomy 32 ( 2011 ) 193 . | in july 2012 , as the four ground - based gamma - ray telescopes of the h.e.s.s .
( _ high energy stereoscopic system _ ) array reached their tenth year of operation in khomas highlands , namibia , a fifth telescope took its first data as part of the system .
this new cherenkov detector , comprising a @xmath0 reflector with a highly pixelized camera in its focal plane , improves the sensitivity of the current array by a factor two and extends its energy domain down to a few tens of gev .
the present part i of the paper gives a detailed description of the fifth h.e.s.s .
telescope s camera , presenting the details of both the hardware and the software , emphasizing the main improvements as compared to previous h.e.s.s . camera technology
. high energy stereoscopic system , h.e.s.s . , camera , electronics , hardware , software , calibration |
a role for modification of activation functions , or intrinsic plasticity ( ip ) , for behavioral learning has been demonstrated for a number of systems @xcite .
for instance , in rabbit eyeblink conditioning , when ion channels related to afterhyperpolarization are being suppressed by a learning event , they can become permanently suppressed .
this has been shown for pyramidal cells of hippocampal areas ca1 and ca3 , and for cerebellar purkinje cells @xcite . in some cases , these changes are permanent and still present after 30 days @xcite , in other cases , intrinsic changes disappear after 3 - 7 days , while the behavioral memory remains intact , raising questions about the long - term component of intrinsic plasticity in these systems .
there are at the present time conflicting ideas on the significance of ip compared to synaptic plasticity @xcite , and the range of functions that ip may have in adaptivity @xcite .
few computational models have been proposed that show how modification in activation functions can be achieved with ion channel based models of realistic single neurons .
marder and colleagues have developed an approach , where they sample a very large parameter space for conductances of ion channels , exploring nonlinearities in the relation between conductances and neural spiking behavior @xcite .
the motivation for this research are observations about neuromodulation and intrinsic plasticity in specific neurons of an invertebrate ganglion ( e.g. , @xcite ) .
they have noted that large variations in some parameters may have little effect on neuronal behavior , while comparatively small variations in certain regions in parameter space may change response properties significantly .
they also suggest that neuromodulation may provide an efficient means of targeting regions in parameter space with significant effects on response properties @xcite .
a study by @xcite assumed the goal of modification of activation functions is to achieve an optimal distribution of firing rates for a population of neurons .
the idea was that by tuning each neuron to a different band of the frequency spectrum , the full bandwidth of frequencies could be employed for information transfer .
this goal was achieved by adjusting @xmath0 , @xmath1 and @xmath2 channels for a generically defined neuron until a desired frequency was stably reached .
we present a different approach , where the modification of activation functions reflects the history of exposure to stimuli for a specific neuron .
similarly , @xcite suggested that synaptic ltp / ltd and linear regulations of intrinsic excitability could operate in a synergistic fashion .
however , in our approach , different types of synaptic stimulation result in state changes for the neuronal unit , influencing its capacity for read - out of stored intrinsic properties .
thus , intrinsic plasticity is conceptualized as fundamentally different from ltp / ltd which does not encompass such a state change .
the learning rule that we derive as the basis for adjustment concerns one - dimensional upregulation or down - regulation of excitability in the `` read - out '' state of the neuron , and affecting only this state .
this rule uses neural activation , significantly determined by intracellular calcium for the learning parameter , which can be shown to be biologically well - motivated ( cf . also @xcite ) .
the membrane voltage @xmath3 is modeled as @xmath4 $ ] .
the individual currents are modeled by conductances , state variables and the reversal potential : @xmath5 the dynamics are defined using state variables for activation ( @xmath6 ) and inactivation ( @xmath7 ) .
the types of equations used for the dynamics are : 1 .
exponential : @xmath8 2 .
logistic : @xmath9 3 .
linexp : @xmath10 the state variables can be defined indirectly using @xmath11 and @xmath12 and one of the equations ( 1 - 3 ) with different values for @xmath13 ( @xmath14 ) , @xmath15 ( @xmath16 and @xmath17 ( @xmath18 , @xmath19 ) .
we use this method for the ion channels in table [ wangbuzsaki ] .
the state variables can also be directly defined ( cf .
@xcite ) : @xmath20 @xmath21 the parameters used are @xmath22 and @xmath23 as in table [ mahon ] .
again , we use one of the equations ( 1 - 3 ) with the @xmath13 parameters ( @xmath24 and @xmath25 ) set to 1 .
( these representations are mathematically equivalent and related by @xmath26 , @xmath27 . )
standard parameter values for the modeling of ion channels ( naive state ) were compared with several publications .
parameter values for @xmath28 , @xmath29 and @xmath30 were adapted from @xcite , for l - type calcium channels ( @xmath31 ) from @xcite and @xcite , see table [ wangbuzsaki ] .
parameters for slow a - type k channels ( @xmath32 ) were adapted from @xcite , for fast a - type k channels ( @xmath33 ) from @xcite , for inward rectifying k channels ( @xmath34 ) from @xcite , and the resulting parameter tables were compared with @xcite and @xcite , see table [ mahon ] .
the maximum conductance of different ion channels can be expressed by a scaling factor in the membrane potential equations as in eq .
[ eq : mu - factorsi ] ( for synaptic currents ) or eq .
[ eq : mu - factorsg ] ( for synaptic conductances ) , cf .
@xmath35 \label{eq : mu - factorsi}\ ] ] @xmath36 \label{eq : mu - factorsg}\ ] ] both nm - independent and nm - dependent modifications may coexist in a neuron , as expressed in eq .
[ eq : mu - kappa - factors ] ( [ nm ] stands for the level of synaptic availability of a neuromodulator nm ) .
@xmath37 \kappa_1 i_1 ) + ( \mu_2 i_2 + [ nm ] \kappa_2 i_2 ) \ldots ] \label{eq : mu - kappa - factors}\ ] ] in this paper , for simplicity , we shall refer to ( eq . [ eq : mu - factorsi ] ) as the generic format for intrinsic adaptation , with the understanding that @xmath38 is replaceable by @xmath39\kappa$ ] .
physiological ranges for @xmath38 can be estimated by various means .
there are measurements for variability in eletrophysiologically defined membrane behavior ( current threshold , spike response to current pulses etc .
@xcite ) that are typically expressed as standard errors ( e.g. 16 - 20% for current threshold , @xcite ) .
there are also attempts at classifying msn cells into different types based on their electrophysiological profile @xcite .
modeling shows that variability of ion channel conductances with a range of @xmath40 matches measures of electrophysiological variability and reproduces the ranges for msn types ( data not shown ) .
interestingly , direct measurements for dopamine d1 receptor - mediated changes on ion channel conductances are approximately in the same ranges ( @xmath41 , @xcite ) .
our discussion is thus based on an estimate of @xmath38 ranging fom 0.6 - 1.4 for each channel .
synaptic input is defined by overlays of the epsps generated by n individual poisson - distributed spike trains with mean interspike interval @xmath42 .
each epsp is modeled as a spike with peak amplitude @xmath43 and exponential decay with @xmath44 similar to @xcite .
ipsps are modeled in a similar way with @xmath45 .
this corresponds to 0.5na ( -0.2na ) as peak current ( with @xmath46 ) .
synaptic conductances are calculated by @xmath47 with @xmath48 set to 0mv .
we have tuned the model to @xmath49= @xmath50 for a first spike for the naive or standard neuron ( all @xmath51 ) . at -40mv ,
this is @xmath52 or @xmath53 , which corresponds to the experimentally measured average value for the rheobase in @xcite .
we may increase the correlation in the input by using a percentage @xmath54 of neurons which fire at the same time .
higher values for @xmath54 increase the amplitude of the fluctuations of the input ( cf .
@xcite ) .
the simulator has been implemented in matlab and executed on an apple 1ghz g4 notebook .
the entire code is interpreted and no specific code optimizations have been applied . for numerical integration , the solver ode45 was used .
simulation of one neuron for 1 s took approximately 68s cpu time .
it is expected that for compiled and optimized code the simulation speed could be increased by at least an order of magnitude .
we explore the impact of small variations in ion channel conductances on the shape of the activation function . as an example , we show the current and conductance changes for a slowly inactivating a - type k+ channel ( kv1.2 , @xmath32 ) , l - type calcium channel ( @xmath31 ) and inward rectifying k+ channel ( @xmath34 ) at different membrane potentials modulated by a scaling factor @xmath55 ( fig .
[ kas - conductance ] , fig . [ actfun ] ) .
regulation of the voltage - dependence @xcite and even of the inactivation dynamics of an ion channel @xcite has also been shown , but these effects are not further discussed here .
we can see that there are critical voltage ranges ( around -50mv , around -80mv and starting at -40mv ) , where the conductance and the current are highest , and where scaling has a significant effect , while scaling has small or no effect in other voltage ranges .
( the na+ current has been disabled for this example to prevent the neuron from firing ) . in fig .
[ temporal ] , we show the current over time - to graphically display the slow dynamics of the @xmath32 and @xmath31 channel . since we do not change the activation - inactivation dynamics of any channel in our model , we show currents only for @xmath56 , @xmath57 = 1 .
we can see that @xmath32 activates moderately fast ( 20ms ) , while it inactivates with a half - time of about 300ms , depending on the voltage . for @xmath31 , activation is almost instantaneous , but inactivation is @xmath58 500ms . the activation function for the msn model shows a time - dependence only in the high - voltage range ( at or above -55mv ) , whereas the components in the lower voltage ranges are not time - dependent .
mathematically , we can consider the individual channels as a set of basis functions that allow function approximation for the activation function .
each particular adjustment of an activation function can be considered learning of a filter which is suited to a specific processing task .
the activation - inactivation dynamics would provide a similar set of basis functions for the temporal domain . of course , it is interesting to note which particular basis functions exist , and also how the temporal dimension is tied in with specific voltage - dependences
. for instance , the slowly inactivating potassium channel @xmath32 provides a skewed mirror image of the function of calcium - gated sk / bk channels , which are responsible for afterhyperpolarization , making different variants of frequency filters possible . on this basis
, a mapping of ion channel components and their density or distributions in different types of neurons could provide an interesting perspective on direct interactions for neurons from different tissue types or brain areas , as well as e.g. between cholinergic interneurons and msns within striatum . to further explore the influence of variability of the activation function
, we apply realistic synaptic input with different amounts of correlation to individual ms neurons ( see fig .
[ currentgsyn ] ) .
this shows us that small adjustments in the contribution of a specific ion channel can result in significantly different spiking behavior even for identical synaptic input .
this occurs when the input is distributed , i.e. has low correlation . in this case
, the neurons spike independently of each other and with different frequencies .
we can eliminate this effect by increasing the correlation of the input .
because of the slow activation / inactivation dynamics of the @xmath32 channel , ( latency of @xmath59 20ms ) only low correlated input activates these channels ( neuronal integrator mode ) , but highly correlated inputs do not activate these channels , driving the membrane to spiking quickly ( coincidence detector mode ) .
therefore correlated input can produce reliable spiking behavior for model neurons which differ in the relative contribution of the slow @xmath32 channel .
distributed input , in contrast , activates slower ion channels , and can produce different tonic firing rates , here according to the contribution of the @xmath32 channels , as long as strong synaptic input keeps the neuron in the relevant voltage range ( persistent activity ) .
similarly , the differential contribution of other channels ( high - voltage gated l - type ca - channels , hyperpolarization - activated girk channels or calcium - dependent sk / bk channels ) will affect neuronal behavior , when the conditions for a prominent influence of these channels are met .
we are modeling a state of ms neurons that exhibits regular tonic firing .
in experimental studies , @xcite showed that ms neurons , similar to cortical neurons , exhibit upstate - downstate behavior , reminiscent of slow - wave sleep , under certain forms of anesthesia ( ketamine , barbiturate ) . however , under neurolept - analgesia ( fentanyl - haloperidol ) , @xcite showed that ms neurons can show driven activity , when cortical input is highly synchronized , and exhibit a state characterized by fluctuating synaptic inputs without rhythmic activity ( i.e. without upstates / downstates ) , when cortical input is desynchronized .
the regular , tonic spiking in this state is very low , much less than in the waking animal , which may be related to the dopamine block by haloperidol .
this makes a waking state of ms neurons characterized by regular tonic spiking at different firing rates probable . in the following ,
we show how intrinsic excitability adaptation can lead to different recalled firing rates under appropriate synaptic stimulation . the model could thus reflect learning that is recalled or read out during msn states under desynchronized cortical input - in contrast to highly synchronized input , which would homogenize the response of the coincidence detecting neurons and favor reliable transfer of spikes .
the general idea for learning intrinsic plasticity is to use a learning parameter @xmath7 for each individual update of the conductance scaling factor @xmath38 .
the _ direction _ of learning ( @xmath60 or @xmath61 ) is determined from the neural activation ( @xmath62 ) for each individual neuron .
neural activation is largely determined by intracellular calcium but here we estimate the neural activation @xmath62 from the spike rate of the neuron , measured over 1 s of simulated behavior ( see [ biolearn ] for a discussion ) .
we define a bidirectional learning rule dependent on an initial firing rate @xmath63 : excitability is increased by a step function @xmath7 ( with stepsize @xmath64 ) when @xmath62 is greater than @xmath63 , excitability is decreased when @xmath62 is lower than @xmath63 ( `` positive learning '' ) .
this means , when the actual neural activation is higher than the initial firing rate , membrane adaptations aim to move the neuron to a higher excitability in order to create a positive memory trace of a period of high activation ( which can then be replicated under distributed synaptic stimulation ) .
the same mechanism applies to lower the excitability of a neuron .
@xmath65 this rule can also be implemented by individual increases in excitability after each action potential , and decreases of excitability for periods of time without action potentials .
initial experiments @xcite indeed show such adaptation of intrinsic excitability after individual spikes . the function @xmath7 can be applied to a single ion channel , such as @xmath32 , but also to a number of ion channels in parallel : e.g. to mimic dopamine d1 receptor activation , @xmath7 may be applied to @xmath56 ( upregulated with high @xmath62 ) , @xmath66 ( downregulated with high @xmath62 ) , and @xmath57 ( downregulated with high @xmath62 ) .
we can show the effect of this learning rule on pattern learning .
we generate synaptic inputs from a grid of 200 input neurons for a single layer of 10 msns . on this grid
we project two stripes of width 4 as a simple input pattern @xmath67 by adjusting the mean interspike interval ( isi ) for the corresponding input neurons to a higher value ( @xmath68 ms for _ on _ vs. @xmath69 ms for _ off _ neurons , see fig .
[ toplo ] ) .
we apply the learning rule to each of the currents @xmath29 , @xmath31 and @xmath32 .
this mimics changes in dopamine d1 receptor sensitivity , which targets these ion channels .
adaptation can be weaker or stronger , depending on learning time ( e.g. , @xmath64=0.01 , @xmath70=20s ( 20 steps ) ( weak ) , @xmath70=40s ( 40 steps ) ( strong ) ) .
after a number of steps , we achieve a distribution of @xmath38-values that reflects the strength of the input ( table [ table3 ] a ) . in fig .
[ positive ] , we obtain spike frequency histograms from the set of ms neurons under different conditions . fig .
[ positive ] a shows the naive response to the input pattern @xmath67 - high activation in two medial areas . after adaptation , this response
is increased ( fig .
[ positive ] b ) . when we apply a test input of a random noise pattern @xmath71 , we see that the learned pattern is still reflected in the spike histogram ( fig .
[ positive ] c ) .
for positive learning , this process is theoretically unbounded , and only limited by the stepsize and the adaptation time .
a saturation state could be defined to prevent unbounded learning , which would also allow to perform capacity calculations .
we should note that applying just one pattern continuously results in a very simple learning trajectory : each update results in a step change in the relevant ion channel currents .
however , we also show that the effects of stepwise adaptation of individual ion channels do not necessarily lead to a completely parallel adaptation of firing rate . in fig .
[ positive ] we see that adaptation is much stronger for high input rather than low input neurons . in this case , @xmath63 at 11.5hz is a fairly low value for neurons to continue to lower their firing rate with stepwise adaptation of the chosen ion channels .
this shows the importance of using appropriate tuning ( harnessing ) mechanisms to make highly nonlinear channels work in a purely linear learning context .
clearly one of the results of learning is an altered spiking behavior of individual neurons dependent on their history .
it is important to realize that this rule is based on neural activation , not synaptic input as a learning parameter - since synaptic input is constant during learning .
we show that this mechanism can be employed not only for positive trace learning , when excitability adaptation corresponds to frequency response , but also for negative trace learning , when excitability adaptation counteracts frequency response and approximates a target firing rate @xmath63 .
this target rate could be set as a result of global inhibitory mechanisms corresponding to the expected mean @xmath72 values under physiological stimulation .
accordingly , the neuron responds with decreases of excitability to high input ranges and increases of excitability to low input ranges ( fig . [ negative ] ) .
@xmath73 this emphasizes that `` homeostatic '' responses - adjusting excitability in the opposite direction to the level of input - can implement trace learning ( pattern learning and feature extraction ) as well .
negative learning rule results in a mirror image of parameter values compared to positive learning , as shown in table [ table3 ] b. the naive response is the same as before ( fig .
[ positive ] a ) . but here , after adaptation , the neurons have habituated to the input , and do not produce a strong response anymore ( fig .
[ neg2 ] a ) . when neurons are tested with @xmath71 , an inverse version of the original pattern appears ( fig .
[ neg2 ] b ) .
similarly , when we apply a different pattern @xmath74 , we obtain a spike histogram , where the learned pattern is overlayed with the new input , resulting in a dampening of the frequency response for @xmath74 ( fig .
[ neg2 ] c ) . for both positive and negative traces ,
learning is pattern - specific , i.e. training with homogeneous , fluctuating ( high - low ) noise , such as @xmath71 , results in no adjustments ( or computes an average ) .
however , any prolonged sequence of neuron - selective stimulation results in neuron - selective patterns .
this requires the population to be protected from prolonged stimulation with random patterns in a biological setting .
we may assume most patterns to be meaningful and highly repetitive , while the neuron exists in a plastic state , while patterns may be random , when the neuron is not plastic ( because it is stimulated with highly correlated or very low frequency input , saturated in its parameters or undergoes ion channel block by selected neuromodulators ) .
a number of experimental results show that intrinsic plasticity in msns may be prominently induced and regulated by intracellular calcium : it has been shown that e.g. the regulation of delayed rectifier @xmath1-channels ( kv2.1 channels ) is effectively performed by @xmath75 influx and calcineurin activation in cultured hippocampal neurons , which can be achieved by glutamate stimulation @xcite .
the regulation concerns marked dephosphorylation ( reduction of conductance ) plus a shift in voltage - dependence .
it has also been shown that 20 s of nmda stimulation , or alternatively , increase of intracellular calcium , increases functional dopamine d1 receptor density at the membrane , which corresponds to an alteration in @xmath76 for d1 parameters , targeting a number of ion channels simultaneously @xcite . for deep cerebellar neurons
, there has recently been some direct evidence on the conditions that induce intrinsic plasticity . here
, alterations in intrinsic excitability can be induced by bursts of epsps and ipsps , accompanied by dendritic calcium transients @xcite . in striatal msns , it has been determined that synaptic stimulation at 1 hz does not cause significant calcium signals , but 10 hz stimulation causes moderate increases , and higher stimulation ( up to 100hz ) significantly raises calcium levels @xcite . in the simulations , neural activation ( @xmath62 )
is estimated from the number of spikes generated , measured over the simulated behavior . in the model case ,
the membrane potential is not used as a separate parameter , because membrane potential and spiking behavior are closely linked . however ,
when a neuron exhibits prominent upstates ( periods of high membrane voltages with a variable number of actual spikes ) , membrane potential may need to be treated as an additional , independent component of @xmath62 , since a great part of the intracellular calcium signal in striatal msns is being generated from high - voltage activated nmda and l - type calcium channels @xcite .
the number of spikes produced nonetheless seems important because of the phenomenon of backpropagating spikes .
backpropagating spikes enhance the calcium signal , thus providing a basis for a prominent role for spiking behavior , or firing rate , for defining intracellular calcium .
the presence of backpropagation of spikes has recently been confirmed for msns @xcite . in general , the induction of intrinsic plasticity may be linked not only to intracellular calcium .
there exists an intricate intracellular system of interactions between diffusible substances like calcium and camp , as well as a number of crucial proteins ( rgs , calcineurin , pka , pkc , other kinases and phosphatases ) for regulating receptor sensitivity and ion channel properties , which are furthermore influenced by nm receptor activation .
thus the learning parameter @xmath7 may be analyzed as being dependent not only on @xmath62 , but also on @xmath39 $ ] , and possibly even a third variable for a - slowly changing - intracellular state .
learning by intrinsic excitability seems particularly suitable for striatal msns , since they have few lateral connections , which provide only a small part of their total input @xcite .
when we have strong recurrent interaction , as in cortex , intrinsic excitability learning needs to adjust activation functions relative to each other , e.g. to ensure optimal distribution of activation functions .
this probably happens in the cortical maps , such as frequency maps in auditory cortex @xcite . in hippocampus ,
synaptic and intrinsic modulation may potentiate each other ( e - s potentiation , @xcite ) , but in other systems ( e.g. striatum ) antagonistic regulation may also exist ( such as ltd combined with positive learning ) , with effects on the balance of synaptic vs. whole - cell localization for the storage of information . when ion channels are regulated by neuromodulation , we can use a factor @xmath39 \kappa$ ] - where @xmath39 $ ] is the extracellular concentration of the ligand and @xmath76 the receptor sensitivity ( see [ variability ] , eq . [ eq : mu - kappa - factors ] ) .
@xmath76 stands for the influence that a nm signal of a certain strength has on a particular ion channel , i.e. the degree of coupling between nm receptor ligand binding and ion channel modification @xcite .
typically , a signal @xmath39 $ ] will regulate several ion channels in parallel , but there may be different @xmath77 for each ion channel . if activation function adaptation proceeds by nm - activated @xmath76 parameters , rather than unconditioned @xmath38 parameters , response to stimuli will consist of an early , non - modulated component , where the input pattern is reflected directly in the spiking frequency , and a later , modulated component , where habituation occurs for a learned pattern , or the stored pattern is reflected by overlaying a new stimulus and the stored pattern .
nm signals orchestrate both adjustments in activation function and synaptic input , with nm activation often depressing synapses , but increasing the variability in the activation function through selected conductance changes ( activating @xmath76-parameters ) . as a result , the input component of the signal is reduced in comparison to the stored intrinsic component after nm activation . presumably , this has a dynamic component , such that for a short time after a strong signal there is an input - dominant phase which is then followed by an intrinsic - dominant phase .
there are different ideas at the present time what intrinsic plasticity can achieve within a network model of neuronal interaction .
a recent review of intrinsic plasticity @xcite has left the authors wondering , whether ip acts mainly to maintain homeostasis , adapting to changes in synaptic strength by keeping neurons within certain ranges but without significant informational capacity , as in the model of @xcite . however , as we have shown , homeostatic adaptation does not exclude information storage under conditions of conditional read - out .
the synergy between synaptic and intrinsic plasticity may take different forms , beyond e - s potentiation .
in contrast to @xcite , @xcite have listed many possible functions and roles of intrinsic adaptive plasticity , based on a review of experimental evidence in different systems .
we have greatly simplified the exposition here by concentrating on spike frequency as a major indicator of neural behavior .
certainly the type of firing ( e.g. burst firing ) is also under control of neuromodulators , and may be influenced by the distribution and density of ion channels .
single neuron computation is more complex than what can be shown with a single compartment model . in dendritic computation
, the coupling of different compartments may be prominently affected by intrinsic plasticity .
for instance , @xcite showed a loss of clustering for k+ channels on the membrane , induced by high glutamate stimulation , indicating a possible input - dependent regulation of dendritic integration .
a recent study on concurrent simulation of synaptic coupling parameters and intrinsic ion channel conductances has concluded that intrinsic and synaptic plasticity can achieve similar effects for network operation @xcite .
we have suggested that synaptic and intrinsic plasticity can substitute for each other , and furthermore that this essential functional parallelism could be an indication for _ information flow _ over time from one modality to the other @xcite .
the direction of this information flow may be from intrinsic to synaptic for the induction of permanent , morphological changes ( such as dendritic spine morphology ) - however the results of @xcite in cerebellum have also shown the possibility of permanent intrinsic plasticity ( albeit in purkinje cells which lack nmda receptors and thus may be highly atypical neurons concerning learning properties ) .
clearly the interaction between synaptic and intrinsic plasticity is still an open question .
here we have shown a simple , local learning mechanism for intrinsic plasticity that allows to store pattern information without synaptic plasticity .
this is different from theoretical approaches , where activation functions are only being modulated to optimize global measures of information transmission between neurons while the information is exclusively stored in synaptic weights .
further work will be needed to investigate the smooth integration of synaptic and intrinsic plasticity and their respective functions in different systems .
we wanted to show quantitatively that ip can have significant effects on spike frequency , dependent on the statistical structure of the input .
in particular , low correlated input , or input during sensitive ( high - voltage membrane ) states induces the strongest variability of spike responses for different activation functions , while highly correlated input acts as drivers for neurons , eliminating subtle differences in activation function . we suggested that starting from a very general , natural format for a learning rule , which can be biologically motivated , we arrive at simple pattern learning , the basis for feature extraction , and realistic types of neural behavior : population - wide increases / decreases of neural firing rates to novel input stimuli , habituation to known stimuli and history - dependent distortions of individual stimuli . a significant application of this theoretical model exists in the observation of pervasive whole - cell adaptations in selected ion channels ( @xmath29 , @xmath31 ) after cocaine sensitization @xcite , with implications of the type of learning that underlies addiction .
this would reduce the dynamic range of intrinsic plasticity .
potentially , then , learning in striatum is mediated in part by intrinsic plasticity , and a reduction in inducible intrinsic plasticity or dynamic range of intrinsic plasticity after cocaine sensitization may contribute to the pathology of addiction .
bargas j , howe a , eberwine j , cao y , surmeier dj ( 1994 ) cellular and molecular characterization of ca++ currents in acutely isolated , adult rat neostriatal neurons
. journal of neuroscience 14 ( 11 ) , 66676686 .
gruber aj , solla sa , surmeier dj , houk jc ( 2003 ) modulation of striatal single units by expected reward : a spiny neuron model displaying dopamine - induced bistability .
journal of neurophysiology 90 , 1095114 .
hu xt , basu s , white fj ( 2004 ) repeated cocaine administration suppresses hva - ca2 + potentials and enhances activity of k+ channels in rat nucleus accumbens neurons .
journal of neurophysiology 92 , 1597607 .
mahon s , deniau jm , charpier s , delord b ( 2000 ) role of a striatal slowly inactivating potassium current in short - term facilitation of corticostriatal inputs : a computer simulation study .
journal of experimental psychology learning memory and cognition 7 , 35762 .
misonou h , mohapatra dp , park ew , leung v , zhen d , misonou k , anderson ae , trimmer js ( 2004 ) regulation of ion channel localization and phosphorylation by neuronal activity .
nature neuroscience 7 , 7118 .
nisenbaum es , mermelstein pg , wilson cj , surmeier dj ( 1998 ) selective blockade of a slowly inactivating potassium current in striatal neurons by ( + /-
) 6-chloro - apb hydrobromide ( skf82958 ) .
synapse 29 , 21324 .
onn sp , fienberg aa , grace aa ( 2003 ) dopamine modulation of membrane excitability in striatal spiny neurons is altered in darpp-32 knockout mice .
journal of pharmacology and experimental therapeutics 306 , 8709 .
schreurs bg , gusev pa , tomsic d , alkon dl , shi t ( 1998 ) intracellular correlates of acquisition and long - term memory of classical conditioning in purkinje cell dendrites in slices of rabbit cerebellar lobule hvi .
journal of neuroscience 18 , 5498507 .
scott l , kruse ms , forssberg h , brismar h , greengard p , aperia a ( 2002 ) selective up - regulation of dopamine d1 receptors in dendritic spines by nmda receptor activation .
proceedings of the national acadadamy of the sciences
u s a 99 , 16614 .
thompson lt , moyer jrj , disterhoft jf ( 1996 ) transient changes in excitability of rabbit ca3 neurons with a time course appropriate to support memory consolidation .
journal of neurophysiology 76 , 183649 .
zhang xf , cooper dc , white fj ( 2002 ) repeated cocaine treatment decreases whole - cell calcium current in rat nucleus accumbens neurons .
journal of pharmacology and experimental therapeutics 301 , 111925 .
= \{0.6 ... 1.4 } ) for the slowly inactivating @xmath1-channel ( kv1.2 , @xmath32 ) , the l - type calcium channel ( @xmath31 ) , and the inward rectifying k+ channel ( @xmath34 ) are shown at different membrane voltages @xmath3 ( a ) in an i - v plot , ( b ) as variability in conductance.,title="fig:",scaledwidth=75.0% ] = \{0.6 ... 1.4 } ) for @xmath32 , @xmath31 , and @xmath34 as components of the activation function ( @xmath78 vs. @xmath3 ) .
the activation function is defined as the membrane voltage response for different injected ( synaptic ) conductances ( @xmath78 ) , and computed by solving eq[eq : mu - factorsg ] for the membrane voltage @xmath3.,title="fig:",scaledwidth=50.0% ] ( b ) the l - type ca channel @xmath31 , and ( c ) for the set of ion channels used in the standard msn model .
we see a rise time due to @xmath32 and overlapping inactivation dynamics in the -55 to -40 mv range.,title="fig:",scaledwidth=75.0% ] neurons with independent poisson processes using different correlations parameters @xmath79 ( a , b ) .
three slightly different neurons with @xmath80 are shown under both conditions .
( a ) response variability and different firing rates for each neuron ( here : 20,26,40 hz ) occur with distributed ( low correlation ) input .
( b ) highly correlated input produces reliable spiking and by implication a single firing rate ( 20hz ) .
the upper panel shows the membrane voltage , the middle panel shows the membrane conductances , and the lower panel shows the synaptic input as conductance.,title="fig:",scaledwidth=75.0% ] = 11.5hz ) ( a ) response of naive neurons to @xmath67 ( b ) response of @xmath67-adapted neurons to @xmath67 ( c ) @xmath67-adapted neurons tested with @xmath71 .
average synaptic input ( @xmath81 ) for each neuron is shown on top .
responses in ( a ) and ( b ) to the same input @xmath67 are different , a pattern similar to @xmath67 emerges in response to uniform ( noise ) pattern input in ( c ) . ,
title="fig:",scaledwidth=75.0% ] = 11.5hz ) ( a ) habituation for neurons adapted to @xmath67 , ( b ) an inverse pattern for @xmath67-adapted neurons tested with @xmath71 and ( c ) interference ( dampening of response ) for a new pattern @xmath74 ( naive : line - drawn bars , adaptive : filled bars ) .
average synaptic input ( @xmath81 ) for each neuron is shown on top .
( a ) shows a uniform response to patterend synaptic input and ( b ) a patterned response to uniform ( noise ) input .
( c ) shows a difference of response for naive vs. @xmath67-adapted neurons to a new pattern @xmath74 .
[ neg2 ] , title="fig:",scaledwidth=75.0% ] | we present an unsupervised , local activation - dependent learning rule for intrinsic plasticity ( ip ) which affects the composition of ion channel conductances for single neurons in a use - dependent way .
we use a single - compartment conductance - based model for medium spiny striatal neurons in order to show the effects of parametrization of individual ion channels on the neuronal activation function .
we show that parameter changes within the physiological ranges are sufficient to create an ensemble of neurons with significantly different activation functions .
we emphasize that the effects of intrinsic neuronal variability on spiking behavior require a distributed mode of synaptic input and can be eliminated by strongly correlated input .
we show how variability and adaptivity in ion channel conductances can be utilized to store patterns without an additional contribution by synaptic plasticity ( sp ) .
the adaptation of the spike response may result in either `` positive '' or `` negative '' pattern learning . however , read - out of stored information depends on a distributed pattern of synaptic activity to let intrinsic variability determine spike response .
we briefly discuss the implications of this conditional memory on learning and addiction . * _ keywords : _ * single neuron model , striatal medium spiny neuron , ion channels , activation function , unsupervised learning , intrinsic plasticity , pattern learning , neuronal variability |
in an attempt to understand the molecular content and physical characteristics of interstellar gas in the low galactic halo , we exploit infrared and ultraviolet data from two nasa satellites : the _ infrared astronomical satellite _ ( _ iras _ ) mission of 1983 and the _ far ultraviolet spectroscopic explorer _ ( _ fuse _ ) satellite of 19992005 . the combination of infrared emission and ultraviolet absorption along sight lines to 45 active galactic nuclei ( agn ) allows us to correlate the infrared cirrus emission intensity with the molecular hydrogen ( h@xmath1 ) absorption column density at select locations at high galactic latitude . _
iras _ mapped the sky in four infrared bands centered on 12 , 25 , 60 , and 100 @xmath0 m .
low et al . ( 1984 ) introduced one of the most surprising results from the _ iras _
maps : diffuse filamentary dust clouds that pervade our galaxy at high latitudes , even in the direction of the galactic poles .
these infrared cirrus " clouds are most prominent at long wavelengths , 100 @xmath0 m , but some can be seen in the 60 @xmath0 m , 25 @xmath0 m , and 12 @xmath0 m bands .
comparisons between _ iras _ dust maps and maps of 21 cm emission reveal a generally good correlation between neutral hydrogen ( hartmann & burton 1997 ) and cirrus dust filaments ( figure 1 ) . because molecular hydrogen ( h@xmath1 ) forms catalytically on the surface of dust grains ( hollenbach , werner , & salpeter 1971 ) , with significant formation rates for grain temperatures @xmath16 k and gas temperatures @xmath17 k ( hollenbach & mckee 1979 ; shull & beckwith 1982 ) , the cold , dusty conditions of the infrared cirrus clouds are favorable for the formation of h@xmath1 .
thus , it is plausible that some fraction of the hydrogen atoms in the cirrus clouds are bound into molecules .
previously , the presence of h@xmath1 in infrared cirrus was inferred indirectly .
first , under the assumption that the infrared emission and total hydrogen column density , n@xmath3 , are proportional , regions of high dust/ ratio , termed infrared excess " , were attributed to the presence of h@xmath1 ( de vries , heithausen , & thaddeus 1987 ; desert , bazell , & boulanger 1988 ; reach , koo , & heiles 1994 ; moritz et al . 1998 ; schlegel , finkbeiner , & davis 1998 , hereafter sfd98 ) .
second , the detection of co in dense cirrus clouds suggests that the diffuse cirrus clouds should contain h@xmath1 as well .
weiland et al .
( 1986 ) compared co maps from magnani , blitz , & mundy ( 1985 ) to _ iras _ maps of infrared cirrus .
each of the 33 co clouds had a cirrus counterpart with similar morphology .
this work established that at least some of the infrared cirrus cloud cores contains co gas .
unfortunately , there is currently no experiment that can map diffuse h@xmath1 emission , either in the 2.12 @xmath0 m [ ( 10 ) s(1 ) ] vibrational line or in the s(0 ) , s(1 ) , s(2 ) pure rotational lines at 28 @xmath0 m , 17 @xmath0 m , and 12 @xmath0 m , respectively .
although h@xmath1 is over @xmath18 times more abundant than co , the ultraviolet and infrared fluorescent emission of h@xmath1 is very weak .
ultraviolet absorption - line spectroscopy is therefore the primary means for detecting cold h@xmath1 in diffuse clouds .
however , it requires background sources with sufficient uv flux to provide adequate signal - to - noise ratio ( s / n ) to detect the weak h@xmath1 lines .
the first major project to conduct such observations was the _ copernicus _ mission of the 1970s ( spitzer & jenkins 1975 ) . however , its sensitivity limited the possible background sources to early - type stars within about 500 pc of the sun .
most ob stars that fit this criterion are at low galactic latitude , and they suffer from confusion and dust extinction in the galactic plane .
individual features in the infrared cirrus can not be discerned at low galactic latitudes , and these stellar sight lines are not effective probes of the dusty filaments .
the _ fuse _ satellite , which has been observing the ultraviolet sky since 1999 , has expanded the opportunities for detecting h@xmath1 .
the increased sensitivity of _ fuse _ ( @xmath19mag ) over _ copernicus _ ( @xmath20mag ) allows us to use more distant stars as well as active galactic nuclei ( agn ) as background sources .
our _ fuse _ survey of h@xmath1 toward high - latitude agn ( gillmon et al.2005 ) is particularly well suited for probing h@xmath1 in infrared cirrus .
the high - latitude sight lines avoid the confusion of the galactic disk , and they provide long path lengths through the galactic halo .
in addition , the random distribution of agn on the sky samples a range of infrared cirrus emission intensities . the main limitation of using pencil - beam " ( absorption ) sight lines to detect h@xmath1 in infrared cirrus is the inability to determine whether the gas and dust detected along a given sight line are physically associated .
therefore , we must rely on indirect correlations between cirrus and h@xmath1 absorption . in this paper
, we compare the h@xmath1 column densities in the survey by gillmon et al .
( 2005 ) with the corresponding infrared cirrus fluxes ( sfd98 ) . by establishing a correlation between the two ,
we assert that at least some of the detected h@xmath1 resides in the cirrus clouds . in
2 we describe the data acquisition and analysis for both _ iras _ and _ fuse_. in 3
we compare the cirrus emission and h@xmath1 absorption and discuss the correlation of the two .
exploiting this correlation and summing over the distribution of h@xmath1 column densities with 100 @xmath0 m cirrus intensity , we estimate the total h@xmath1 mass ( @xmath21 ) in cirrus clouds around the milky way . with molecular fractions ranging from 130% , the total cirrus mass throughout the milky way
is @xmath22 . in
4 we summarize our results and the implications of finding this amount of gas in the low halo of the milky way .
the h@xmath1 absorption data were taken from _ fuse _ spectra of agn , using standard data analysis techniques ( tumlinson et al . 2002 ; gillmon et al .
studies of h@xmath1 have been a major part of the _ fuse _ science plan .
the satellite , its mission , and its on - orbit performance are described in moos et al .
( 2000 ) and sahnow et al .
scientific results on interstellar h@xmath1 have appeared in a number of papers ( shull et al .
2000 ; snow et al . 2000 ; rachford et al . 2002 ; richter et al .
2001 , 2003 ; tumlinson et al . 2002
; shull et al .
the resolution of _ fuse _ varies from @xmath23 across the far - uv band .
all observations were obtained in time - tag ( ttag ) mode , using the @xmath24 lwrs aperture , with resolution @xmath25 km s@xmath8 at 1050 .
the s / n of the co - added data ranges from 211 per pixel ; the s / n per resolution element varies with spectral resolution , which is not fixed in our survey .
most of the data were binned by 4 pixels before analysis , with the rare case of binning by 2 or 8 pixels .
the data in this paper were taken from our high - latitude h@xmath1 survey ( gillmon et al .
2005 ) , which describes our search for h@xmath1 absorption along 45 sight lines to background agn at galactic latitudes @xmath26 . the 45 agn in the survey by gillmon et al .
( 2005 ) were a subset of the 219 _ fuse _ targets selected in wakker et al .
( 2003 ) as candidates for the analysis of galactic ovi .
these targets probe diffuse gas in both the local galactic disk and low galactic halo . of the available agn at high latitude ,
45 sight lines were chosen , based on an imposed s / n requirement of @xmath27 or @xmath28 with 4-pixel binning .
the observed h@xmath1 lines arise from the lyman and werner electronic transitions , from the ground electronic state , x @xmath29 , to the excited states , b @xmath30 ( lyman bands ) and c @xmath31 ( werner bands ) .
the rotational - vibrational lines arise from the ground vibrational state and a range of rotational states .
our analysis was restricted to ten vibrational - rotational bands , lyman ( 00 ) to ( 80 ) and werner ( 00 ) , which are located between 1000 @xmath32 and 1126 @xmath32 .
the vibrational state notation is ( @xmath33 ) . in most sight lines ,
we observed absorption lines from rotational states @xmath343 , and sometimes up to @xmath35 .
the end product of the h@xmath1 absorption - line analysis is the column density , n@xmath11 ( @xmath36 ) , the physical density of h@xmath1 molecules , integrated along the sight line .
each absorption line was fitted with a voigt profile in order to determine the equivalent width , a measure of the absorbed light in the line .
the equivalent widths were fitted to a curve of growth , to find the column density , n@xmath37 in each rotational state @xmath38 .
the points for each @xmath38 were tied together during the fitting to produce a single , consistent column density for each rotational state , n(@xmath38 ) .
the sum of the column densities in all rotational states then gives the total column density , n@xmath11 .
fuse spectra of 87% ( 39 of 45 ) of the observed agn showed detectable h@xmath1 absorption , with column densities ranging from n@xmath11 @xmath39 @xmath36 .
the _ fuse _ survey is sensitive to n@xmath11 @xmath40 @xmath36 , depending on the s / n ( 211 per pixel ) and spectral resolution ( @xmath4120,000 ) . to obtain infrared cirrus emission intensities ( mjy sr@xmath8 ) , we use the 100 @xmath0 m maps presented by sfd98 .
these were a composite of data from the _ iras _ mission of 1983 and the _ cobe _ mission of 19891990 , capitalizing on the strengths of each .
because the interstellar dust emits like a grey body " , the emission intensity is sensitive to the dust temperature . as a result ,
two regions with the same dust column density but different dust temperatures will have different infrared intensities .
to correct for this effect , sfd98 used the ratio of the 100 @xmath0 m and 240 @xmath0 m _ cobe _ maps to produce a map of dust color temperature .
they used this ratio to correct the 100 @xmath0 m _ iras _ map so that it is proportional to dust column density .
_ iras _ mapped the sky in four broadband infrared channels , centered at 12 , 25 , 60 , and 100 @xmath0 m with a resolution of @xmath42 , while _ cobe _ mapped the sky in 10 broad photometric bands from 1 to 240 @xmath0 m at a resolution of @xmath43 . before combining the data sets , sfd98 took great care in the difficult removal of the zodiacal foreground emission and _ iras _ striping artifacts that arise from differences in solar elongation between scans . confirmed point sources
were also removed .
the maps were combined in such a way as to preserve the _ cobe _
calibration and _ iras _ resolution .
maps of the temperature - corrected 100 @xmath0 m intensity , @xmath44 , are presented for the northern galactic hemisphere ( figure 2 ) and for the southern galactic hemisphere ( figure 3 ) .
we overplot the locations of the 45 sight lines from the _ fuse _
h@xmath1 survey : 28 northern agn and 17 southern agn .
before the _ fuse _ mission , the direct detection of h@xmath1 in infrared cirrus by uv absorption - line spectroscopy was prevented mainly by a lack of background sources at high galactic latitude . even though this problem has been alleviated by _
s ability to observe bright agn as background sources , another problem with absorption - line spectroscopy along pencil - beam " sight lines comes to the forefront .
absorption studies lack morphological information and can not provide distances to clouds along the line of sight .
therefore , it is difficult to identify the gas that gives rise to the detected column density . the infrared cirrus problem is a classic case . for any given sight line ,
it is possible that the h@xmath1 absorption is not physically associated with the infrared cirrus along the beam .
if this is the case , then comparing n@xmath11 with the cirrus dust column density would lead to erroneous results .
it would be helpful to determine whether the diffuse h@xmath1 along all sight lines resides in the cirrus .
if no other component of the diffuse ism harbored significant amounts of h@xmath1 , then n@xmath11 for a given sight line could safely be associated with other cirrus properties . in this section
, we investigate this possibility , based on a property of h@xmath1 called self - shielding " .
following line absorption of uv photons from the mean interstellar radiation field , h@xmath1 decays approximately 11% of the time to the dissociative continuum . as a result ,
the molecular fraction , @xmath46 is generally larger in clouds with a greater total hydrogen column density , n@xmath3 = n@xmath47 + 2n@xmath11 .
molecules on the outside of the cloud shield those in the interior from dissociating uv ( hollenbach , werner , & salpeter 1971 ; black & dalgarno 1976 ; browning , tumlinson , & shull 2003 ) . in optically thin clouds
, the density of molecules can be approximated by the equilibrium between formation and destruction , @xmath48 in this formula , the numerical value for @xmath2 is scaled to fiducial values of hydrogen density , @xmath49 ( 30 @xmath10 ) , h@xmath1formation rate coefficient , @xmath50 ( @xmath51 @xmath52 s@xmath8 ) , and mean h@xmath1 pumping rate in the fuv lyman and werner bands , @xmath53 s@xmath8 .
the h@xmath1 photodissociation rate is written as @xmath54 , where the coefficient @xmath55 is the average fraction of fuv excitations of h@xmath1 that result in decays to the dissociating continuum .
the h@xmath1 formation rate per unit volume is written as @xmath56 , where the coefficient @xmath50 depends on the gas temperature , grain surface temperature , and gas metallicity ( @xmath57 ) .
the metallicity dependence comes from the assumed scaling of grain - surface catalysis sites with the grain / gas ratio . for sight lines in the local galactic disk ,
this rate coefficient has been estimated ( jura 1974 ) to range from @xmath58 @xmath52 s@xmath8 at solar metallicity . this standard value for @xmath50
is expected to apply at suitably low temperatures of gas ( @xmath59 k ) and grains ( @xmath60 k ) as discussed by shull & beckwith ( 1982 ) and hollenbach & mckee ( 1979 ) .
the effects of self - shielding were observed by plotting @xmath2 versus n@xmath3 , the total column density of hydrogen .
as h@xmath1 absorption lines in the lyman and werner bands become optically thick , the rate ( @xmath61 ) of uv pumping and molecular dissociation diminish . in this way , the presence of h@xmath1 screens molecules in the inner portions of the cloud from dissociation . a transition from low ( @xmath62 ) to high ( @xmath63 ) molecular fractions at n@xmath64 @xmath36
was noted in the _ copernicus _
h@xmath1 survey ( savage et al . 1977 ) .
a recent _ fuse _ survey of galactic disk stars ( shull et al .
2005 ) provided similar results ( figure 4 ) . in assessing the molecular content of the cirrus
, we consider three generic cases .
the first possibility ( case i ) is that all the detected h@xmath1 resides in cirrus clouds , with no contribution from foreground clouds . under the assumption that the total hydrogen density , @xmath65 , is proportional to the number density of dust grains in
any given interstellar cloud ( hollenbach & mckee 1979 ) , the cirrus dust column density should be proportional to the amount of n@xmath66 associated with h@xmath1 . a plot of n@xmath11 vs. dust column density
should then show the familiar self - shielding transition of molecular hydrogen . in case
ii , some of the observed h@xmath1 lies in the cirrus , but some resides in another component of the ism not visible in the cirrus maps .
a plot of n@xmath11 vs. dust column density would show some points that follow the self - shielding transition ( those corresponding to cirrus ) .
however , the transition would be broadened by points that are randomly distributed .
if a significant portion of the detected h@xmath1along a sight line is not in cirrus , n@xmath11 will not correlate with dust column density .
this effect would produce sight lines with low dust column density and significant amounts of h@xmath1 .
the extreme situation is case iii , in which none of the h@xmath1 resides in cirrus clouds .
we consider this unlikely , given the number of observed regions with infrared excess " ( high dust / n@xmath47 ratios ) and the detection of co in denser cirrus clouds .
once the n@xmath11 in cirrus exceeds @xmath67
@xmath36 , it should be detectable by _
fuse_. for case iii to be consistent with our survey , the covering fraction of cirrus regions with detectable h@xmath1 would need to be quite small , whereas the cirrus covering factor is observed to be @xmath5% at @xmath6 ( see 3.4 ) .
the issue then comes down to whether case i or case ii is a better description of the cirrus - h@xmath1 correlation .
figure 5 presents a plot of n@xmath11 along sight lines to the 45 agn in the _ fuse _ survey vs. the temperature - corrected ir flux at each location from the sfd98 maps ( blue diamonds ) .
if the temperature - corrected flux is proportional to dust column density , this plot can be used to test the three cases mentioned above .
the blue diamonds appear to show the self - shielding transition discussed in case i , implying that much of the detected h@xmath1 is in the cirrus . in this section , we describe five additional sight lines shown in red ( figure 5 ) toward the regions of lowest cirrus .
these additional sight lines were chosen to explore whether the correlation between n@xmath11and 100 @xmath0
m cirrus is universal .
table 5 of sfd98 lists the coordinates of the regions of lowest ( temperature - corrected ) 100 @xmath0 m intensity , @xmath44 , from their maps .
we searched the _ fuse _ archive for targets within @xmath68 of the given coordinates and found five targets behind regions of low cirrus ( @xmath69 mjysr@xmath8 ) that also had _ fuse _
data with sufficient ( s / n)@xmath70 to conduct an h@xmath1 analysis .
these five targets are in addition to those in the gillmon et al .
( 2005 ) sample of 45 agn .
they are listed in table 1 and shown as asterisks in figure 5 .
four of these five targets showed no evidence of h@xmath1 , with a typical upper limit of n@xmath11 @xmath71 @xmath36 .
this result lends further credence to the theory that most of the observed h@xmath1 is in the cirrus ( case i )
. however , one target , ugc5720 , showed significant h@xmath1 absorption lines ( figure 6 ) .
an analysis , as described in 2.1 , yielded a significant column density n@xmath11 @xmath72 @xmath36 . in 3.3 ,
we explore possible explanations for this anomalous sight line and for the spread in the self - shielding transition .
figure 5 plots n@xmath11 vs. temperature - corrected intensity , @xmath44 ( mjysr@xmath8 ) , and exhibits a clear self - shielding transition of h@xmath1 .
this indicates that a significant fraction of the detected h@xmath1 resides in the infrared cirrus .
however , with small - number statistics on 50 agn sight lines , the transition is not sharp , occurring between log @xmath44 = 0.20.5 .
there is one outlying point ( ugc5720 ) with low cirrus intensity , log @xmath73 , but a significant column density , log n@xmath11 @xmath74 .
this outlier suggests that not all the detected h@xmath1 resides in cirrus clouds , and that other components of the diffuse gas may harbor detectable amounts of molecules . in some sight lines that intercept non - cirrus clouds with low 100 @xmath0 m intensity , the gas may undergo an early molecular transition and appear with higher n@xmath11
there are other explanations that might produce the same effects , even if all the h@xmath1 resides in cirrus clouds .
one consideration is how well the temperature - corrected flux map traces the dust column density .
to correct for temperature variations , sfd98 created a temperature map from the ratio of the _ cobe _ 100 @xmath0 m and 240 @xmath0 m maps .
thus , low - resolution ( 1.1@xmath75 ) temperature maps were used to correct the @xmath76 resolution _ iras _ map to produce a 100 @xmath0 m map at @xmath77 resolution .
it is likely that the dust temperature varies on smaller scales than can be resolved by this method , and a more accurate temperature correction might tighten the self - shielding transition .
it is also possible that ugc5720 lies behind a region of cold cirrus dust unresolved by the temperature map .
thus , a significant portion of the dust column density might not be indicated by the observed flux .
a related consideration is that the _ iras _ map might not resolve significant variations in dust column density .
the sight lines observed by _ fuse _ absorption probe gas along a very narrow beam , while the _ iras _
beam is considerably larger .
thus , if there were significant structure on smaller scales than the _ iras _ beam , _ fuse _ could observe a denser clump of h@xmath1 while the _ iras _
emission would be beam - diluted " .
a higher resolution infrared map might tighten the self - shielding transition and/or bring the outlier ugc5720 onto the correlation . figure 7 shows a @xmath78 section of the sfd98 flux map centered on ugc5720 .
large fluctuations in temperature - corrected flux near the position of the agn suggest that the sight line may be picking up a clump of cirrus with higher dust column density unresolved by these maps .
finally , there is the possibility that there are multiple cirrus features superimposed along a line of sight .
duel & burton ( 1990 ) compared the morphology of cirrus clouds and h i maps in various velocity intervals to show that cirrus features that appear simple are , in some cases , superpositions of kinematically distinct components .
this idea of a concatenation of clouds " was proposed for translucent h@xmath1 clouds ( browning et al .
2003 ) to explain the high levels of h@xmath1 rotational excitation in systems with large n@xmath11 .
these authors suggested that , if a sight line intercepts multiple , physically distinct cloud components , the h@xmath1 will be exposed to a radiation field enhanced over that expected for a single , contiguous cloud with the same total column density .
the enhanced h@xmath1 destruction rate from the stronger uv radiation would reduce the mean molecular fraction and produce a gradual transition to higher n@xmath11 .
some sight lines with seemingly high dust column density , but low n@xmath11 , may actually be probing multiple , superimposed filaments with high integrated dust column density . with the empirical correlation ( figure 5 ) between ir cirrus and h@xmath1 column density , we can make a quantitative estimate of the h@xmath1 mass in the diffuse cirrus clouds .
we begin with the _ iras _
maps of the northern galactic hemisphere .
we assume that these cirrus clouds lie at elevation @xmath79 above the milky way plane , and that clouds of intensity @xmath80 cover a fraction @xmath81 of the planar area at @xmath82 .
the total h@xmath1 mass in this planar cloud deck is then , @xmath83 \left [ \frac { \langle f_c \rangle } { 0.5 } \right ] z_{100}^2 \ ; , \ ] ] where n@xmath11@xmath84 is the mean h@xmath1 column density corresponding to cirrus intensity @xmath80 ( figure 5 ) and where @xmath85 is the radius of the cirrus disk , at elevation @xmath86 subtended by the cone at @xmath87 . for this estimate
, we have assumed that the cirrus covers a fraction @xmath88 of the sky at @xmath82 , independent of intensity .
we now make a more careful calculation , summing over the actual data from _ iras _ and _ fuse_. figure 8 shows the distribution of cirrus covering factors , @xmath81 , averaged over northern hemisphere regions with @xmath89 greater than @xmath90 , @xmath91 , and @xmath92 , respectively . for our fiducial cone at @xmath82 ,
approximately 50% of the sky is covered by cirrus with intensity log @xmath93 , the value corresponding to the h@xmath1 transition ( figure 5 ) .
performing the full summation ( eq .
3 ) over 8 logarithmic intensity bins above the transition , of width @xmath94(log @xmath80 ) = 0.1 , between log @xmath95 and 1.0 , we find a total h@xmath1 mass of @xmath96 contained in the cirrus at @xmath82 . to convert this calculation to the inner milky way , within the solar circle , we multiply by a factor @xmath97 , for the northern and southern galactic hemispheres , and scale by a factor @xmath98 to account for the number of similar conical areas around the galactic disk . note that this total h@xmath1 mass is independent of the assumed cirrus elevation , @xmath86 , since the area - scaling cancels the factor @xmath99 in equation ( 3 ) .
we arrive at an extrapolated total molecular mass over the inner milky way , @xmath100 , assuming that the cirrus along the agn sightlines at @xmath6 is typical .
the molecular fractions of these cirrus clouds , with n@xmath11 @xmath101 @xmath36 , range from 130% for log @xmath44 = 0.20.5 ( see figure 6 of gillmon et al .
2005 ) , with an average @xmath102 for clouds with log n@xmath11 @xmath103 .
therefore , we estimate the total gas mass in the cirrus to be @xmath104 .
the characteristics of the cirrus clouds along our agn sight lines can be verified by computing the dust - to - gas " ratio , defined as the ratio of 100 @xmath0 m cirrus intensity , @xmath44 ( mjy sr@xmath8 ) , to h i column density , n@xmath47 ( @xmath36 ) .
table 2 gives these values and their ratio for 16 of our agn sight lines observed in 21 cm ( lockman & savage 1995 ) .
this ratio ranges from @xmath105 mjy sr@xmath8 @xmath106 toward pg 0953 + 414 to @xmath107 mjy sr@xmath8 @xmath106 toward 3c 273 .
the mean value and standard deviation are @xmath108 mjy sr@xmath8 @xmath106 , when we exclude the anomalous sight line to 3c 273 , which lies behind radio loop i and the north polar spur .
this mean ratio is in excellent agreement with the mean value , @xmath109 mjy sr@xmath8 @xmath106 , found by boulanger , baud , & van albada ( 1985 ) in a @xmath110 field at high galactic latitude .
we have undertaken a comparison between the column density , n@xmath11 , and the 100 @xmath0 m cirrus intensity for a total of 50 sight lines . for the cirrus
, we used the temperature - corrected maps of schlegel , finkbeiner , & davis ( 1998 ) , and adopted h@xmath1 column densities from our high - latitude survey ( gillmon et al .
the presence of a clear correlation between uv ( h@xmath1 ) absorption and ir ( cirrus ) emission indicates that a significant fraction of the h@xmath1 is physically associated with the cirrus clouds .
however , the self - shielding transition of h@xmath1 fraction used to define the correlation is not sharp .
the existence of one outlying sight line suggests either that some of the detected h@xmath1 may exist in another component of the diffuse ism , or that the limited resolution of the infrared maps is obscuring the physical conditions .
of the three possible cases for h@xmath1cirrus connections laid out in 3.1 ,
case i or ii best describe the data . put simply , h@xmath1 is contained in most , if not all diffuse cirrus clouds . at galactic latitudes
@xmath6 , approximately 50% of the sky is covered with cirrus , at temperature - corrected 100 @xmath0 m intensities @xmath7 mjy sr@xmath8 . with this correlation , we have found a convenient means of identifying the best extragalactic sight lines for h@xmath1-clean " far - uv absorption studies of intergalactic or interstellar matter .
conversely , if the goal is to study h@xmath1 at the disk - halo interface , the cirrus maps would be a good guide .
we also made a rough estimate of the h@xmath1 mass contained in these cirrus clouds . exploiting the h@xmath1cirrus correlation
, we summed the distributions of ir cirrus intensity and h@xmath1 column density to find @xmath21 in cirrus h@xmath1 and @xmath22 in total hydrogen , distributed over the milky way disk - halo interface , within the solar circle . above the self - shielding transition , these
diffuse halo clouds have molecular fractions ranging from 130% , for column densities n@xmath111 @xmath36 and n@xmath11 @xmath101 @xmath36 ( gillmon et al . 2005 ) . to support such high molecular fractions by h@xmath1 formation on grain surfaces , the cirrus clouds are probably compressed sheets with densities @xmath9 @xmath10 , in which the gas and grains remain sufficiently cold to form h@xmath1 . on average , the ir cirrus clouds may actually have higher molecular fractions , on average , than diffuse clouds in the galactic disk .
this point was also made by reach et al .
( 1994 ) , who estimated that the h@xmath1/h i transition in denser cirrus clouds occurs at n@xmath112 @xmath36 , on the basis of fits to the far - infrared excess .
this transition column density is almost twice that found here , probably because the infrared - excess technique requires larger molecular fractions than used for the uv absorbers ( @xmath113 ) .
this initial survey opens up considerable opportunities for future studies of the h@xmath1cirrus correlation .
higher resolution infrared maps with more accurate temperature corrections , such as those obtainable with the _
spitzer space telescope _
, would greatly improve the effectiveness of this method .
a comparison with h i 21-cm maps ( e.g. , lockman & condon 2005 ) could delineate the gaseous structures associated with the cirrus and measure the dust - to - gas ratios in these diffuse clouds . expanding the uv sample to include more sight lines to background agn
would improve the statistics of such a small sample .
the 45 agn in the survey by gillmon et al .
( 2005 ) were a subset of the 219 _ fuse _ targets selected in wakker et al .
( 2003 ) as candidates for the analysis of galactic ovi .
the next 50 brightest targets have an average flux of @xmath114 erg @xmath36 s@xmath8 @xmath8 . to achieve a s / n of 3 per pixel with _ fuse
_ would require approximately 20 ksec per agn , for a program total of 1000 ksec .
an ultraviolet telescope with sensitivity greater than fuse is probably necessary for feasible exposure times in a @xmath115 target survey . another promising avenue for future exploration is to compare the cirrus maps and h@xmath1 absorption lines with other tracers , such as h i , co and @xmath116-ray emission . on the basis of such comparisons , grenier , casandjian , & terrier ( 2005 ) suggest that many interstellar clouds in the solar neighborhood have extensive dark regions that bridge the dense cloud cores to atomic phases .
these details are beyond the scope of our current paper . perhaps the most direct way to investigate the connection between diffuse h@xmath1 and
ir cirrus would be to map h@xmath1 in uv or ir fluorescent emission .
this method would provide the morphological information lacking in pencil - beam " uv - absorption sight lines .
there is currently no high - resolution experiment that can map diffuse h@xmath1 emission , either in the mid - infrared ( 28 and 17 @xmath0 m ) or the far - ultraviolet ( 10001100 ) .
intriguing results for h@xmath1 ultraviolet fluorescent emission at @xmath117 resolution may soon be available from the _ spectroscopy of plasma evolution from astrophysical radiation _
( spear ) mission ( edelstein et al .
with appropriate ir and uv telescopes , these methods could help map the gaseous galactic halo .
we thank ken sembach and jay lockman for useful discussions .
this work was based in part on data obtained for the guaranteed time team team by the nasa - cnes - csa _ fuse _ mission operated by the johns hopkins university .
financial support to u.s .
participants has been provided by nasa contract nas5 - 32985 .
the colorado group also received _ fuse _ support from nasa grant nag5 - 10948 for studies of interstellar h@xmath1 .
lccccc ms0354 - 3650 & 238.87 & @xmath118 & 0.37 & 3 & @xmath119 + har03 & 152.41 & 52.87 & 0.38 & 2 & @xmath120 + ugc5720 & 156.20 & 52.80 & 0.41 & 3 & @xmath121 + ngc5447 & 102.82 & 59.83 & 0.44 & 2 & @xmath120 + m101d & 102.67 & 59.75 & 0.46 & 2 & @xmath122 lccc & 1.98 & 20.14 & 1.43 + 3c 273 & 1.13 & 19.42 & 4.29 + h 1821 + 643 & 2.34 & 20.34 & 1.07 + hs 0624 + 6907 & 5.29 & 20.62 & 1.27 + mrc 2251 - 178 & 2.14 & 20.12 & 1.62 + mrk 205 & 2.28 & 20.25 & 1.28 + mrk 421 & 0.83 & 19.73 & 1.54 + pg 0804 + 761 & 1.92 & 20.41 & 0.74 + pg 0844 + 349 & 2.00 & 20.24 & 1.15 + pg 0953 + 414 & 0.66 & 20.00 & 0.66 + pg 1116 + 215 & 1.23 & 19.83 & 1.82 + pg 1211 + 143 & 1.86 & 20.33 & 0.87 + pg 1259 + 593 & 0.44 & 19.67 & 0.95 +
pg 1302 - 102 & 2.36 & 20.22 & 1.42 + pks 0405 - 12 & 3.22 & 20.28 & 1.69 + pks 2155 - 304 & 1.19 & 20.06 & 1.03 + | we combine data from our recent _ fuse _ survey of interstellar molecular hydrogen absorption toward 50 high - latitude agn with cobe - corrected _ iras _ 100 @xmath0 m emission maps to study the correlation of infrared cirrus with h@xmath1 .
a plot of the h@xmath1 column density vs. ir cirrus intensity shows the same transition in molecular fraction , @xmath2 , as seen with total hydrogen column density , n@xmath3 .
this transition is usually attributed to h@xmath1 self - shielding " , and it suggests that many diffuse cirrus clouds contain h@xmath1 in significant fractions , @xmath4 130% .
these clouds cover @xmath5% of the northern sky at @xmath6 , at temperature - corrected 100 @xmath0 m intensities @xmath7 mjy sr@xmath8 .
the sheetlike cirrus clouds , with hydrogen densities @xmath9 @xmath10 , may be compressed by dynamical processes at the disk - halo interface , and they are conducive to h@xmath1 formation on grain surfaces . exploiting the correlation between n@xmath11 and 100 @xmath0 m intensity , we estimate that cirrus clouds at @xmath6 contain @xmath12 in h@xmath1 . extrapolated over the inner milky way , the cirrus may contain @xmath13 of h@xmath1 and @xmath14 in total gas mass . if elevated to 100 pc , their gravitational potential energy is @xmath15 erg . |
a variety of data in many different fields can be described by networks .
examples include friendship and social networks , food webs , protein - protein interaction and gene regulatory networks , the world wide web , and many others .
one of the fundamental problems in network science is link prediction , where the goal is to predict the existence of a link between two nodes based on observed links between other nodes as well as additional information about the nodes ( node covariates ) when available ( see , @xcite and @xcite for recent reviews ) .
link prediction has wide applications .
for example , recommendation of new friends or connections for members is an important service in online social networks such as facebook . in biological networks , such as protein - protein interaction and gene regulatory networks ,
it is usually time - consuming and expensive to test existence of links by comprehensive experiments ; link prediction in these biological networks can provide specific targets for future experiments .
there are two different settings under which the link prediction problem is commonly studied . in the first
setting , a snapshot of the network at time @xmath0 , or a sequence of snapshots at times @xmath1 , is used to predict new links that are likely to appear in the near future ( at time @xmath2 ) . in the second
setting , the network is treated as static but not fully observed , and the task is to fill in the missing links in such a partially observed network .
these two tasks are related in practice , since a network evolving over time can also be partially observed and a missing link is more likely to emerge in the future . from the analysis point of view , however , these settings are quite different ; in this paper , we focus on the partially observed setting and do not consider networks evolving over time .
there are several types of methods for the link prediction problem in the literature .
the first class of methods consists of unsupervised approaches based on various types of node similarities .
these methods assign a similarity score @xmath3 to each pair of nodes @xmath4 and @xmath5 , and higher similarity scores are assumed to imply higher probabilities of a link .
similarities can be based either on node attributes or solely on the network structure , such as the number of common neighbors ; the latter are known as structural similarities .
typical choices of structural similarity measures include local indices based on common neighbors , such as the jaccard index @xcite or the adamic - adar index @xcite , and global indices based on the ensemble of all paths , such as the katz index @xcite and the leicht - holme - newman index @xcite .
comprehensive reviews of such similarity measures can be found in @xcite and .
another class of approaches to link prediction includes supervised learning methods that use both network structures and node attributes .
these methods treat link prediction as a binary classification problem , where the responses are @xmath6 indicating whether there exists a link for a pair , and the predictors are covariates for each pair , which are constructed from node attributes .
a number of popular supervised learning methods have been applied to the link prediction problem .
for example , @xcite and @xcite use the support vector machine with pairwise kernels , and @xcite compares the performance of several supervised learning methods .
other supervised methods use probabilistic models for incomplete networks to do link prediction , for example , the hierarchical structure models @xcite , latent space models @xcite , latent variable models @xcite , and stochastic relational models @xcite .
our approach falls in the supervised learning category , in the sense that we make use of both the node similarities and observed links . however , one difficulty in treating link prediction as a straightforward classification problem is the lack of certainty about the negative and positive examples
this is particularly true for negative examples ( absent edges ) . in biological networks
in particular , there may be no certain negative examples at all @xcite .
for instance , in a protein - protein interaction network , an absent edge may not mean that there is no interaction between the two proteins
instead , it may indicate that the experiment to test that interaction has not been done , or that it did not have enough sensitivity to detect the interaction .
positive examples could sometimes also be spurious for example , high - throughput experiments can yield a large number of false positive protein - protein interactions @xcite . here
we propose a new link prediction method that allows for the presence of both false positive and false negative examples .
more formally , we assume that the network we observe is the true network with independent observation errors , i.e. , with some true edges missing and other edges recorded erroneously .
the error rates for both kinds of errors are assumed unknown , and in fact can not be estimated under this framework .
however , we can provide rankings of potential links in order of their estimated probabilities , for node pairs with observed links as well as for node pairs with no observed links .
these relative rankings rather than absolute probabilities of edges are sufficient in many applications .
for example , pairs of proteins without observed interactions that rank highly could be given priority in subsequent experiments .
to obtain these rankings , we utilize node covariates when available , and/or network topology based on observed links .
the rest of the paper is organized as follows . in section [ sec :
model ] , we specify our ( rather minimal ) model assumptions for the network and the edge errors .
we propose link ranking criteria for both directed and undirected networks in section [ sec : meth ] .
the algorithms used to optimize these criteria are discussed in section [ sec : alg ] . in section [ sec : sim ] we compare performance of proposed criteria to other link prediction methods on simulated networks . in section [ sec
: data ] , we apply our methods to link prediction in a protein - protein interaction network and a school friendship network .
section [ sec : summary ] concludes with a summary and discussion of future directions .
a network with @xmath7 nodes ( vertices ) can be represented by an @xmath8 adjacency matrix @xmath9 $ ] , where @xmath10 we will consider the link prediction problem for both undirected and directed networks
. therefore @xmath11 can be either symmetric ( for undirected networks ) or asymmetric ( for directed networks ) . in our framework , we distinguish between the adjacency matrix of the true underlying network @xmath12 , and its observed version @xmath11 .
we assume that each @xmath13 follows a bernoulli distribution with @xmath14 .
given the true network , we assume that the observed network is generated by @xmath15 where @xmath16 and @xmath17 are the probabilities of correctly recording a true edge and an absent edge , respectively . note that we assume that this probability is constant and does not depend on @xmath4 , @xmath5 , or @xmath18 .
then we have @xmath19 if the values of @xmath16 , @xmath17 and @xmath18 were known , then the probabilities of true edges conditional on the observed adjacency matrix could have been estimated as @xmath20 it is easy to check that both and are monotone increasing functions of @xmath18 .
taking into account implies that they are also increasing functions of @xmath21 as long as @xmath22 .
this gives us a crucial observation : if the goal is to obtain relative rankings of potential links , it is sufficient to estimate @xmath21 , and it is not necessary to know @xmath16 , @xmath17 and @xmath18 . an important special case in this setting is @xmath23
. then all the observed links are true positives , and we only need to provide a ranking for node pairs without observed links .
this can be applied in recommender systems , for example , for recommending possible new friends in a social network .
another special case is when @xmath24 , which corresponds to all absent edges being true negatives .
this setting can be used to frame the problem of investigating reliability of observed links , for example , in a gene regulatory network inferred from high - throughput gene expression data .
an estimate of @xmath25 $ ] provides rankings for both these special cases and the general problem , and thus we focus on estimating @xmath21 for the rest of the paper .
in this section , we propose criteria for estimating the probabilities of edges in the observed network , @xmath26 , for both directed and undirected networks .
the criteria rely on a symmetric matrix @xmath27 $ ] with @xmath28 , which describes the similarity between nodes @xmath4 and @xmath29 .
the similarity matrix @xmath30 can be obtained from different sources , including node information , network topology , or a combination of the two .
we will discuss choices of @xmath30 later in this section .
first we consider directed networks .
the key assumption we make is that if two pairs of nodes are similar to each other , the probability of links within these two pairs are also similar .
specifically , in figure [ fig : directed - intuition ] , @xmath18 and @xmath31 are assumed close in value if node @xmath4 is similar to node @xmath29 and node @xmath5 is similar to node @xmath32 . for directed networks , we measure similarity of node pairs @xmath33 and @xmath34 by the product @xmath35 ( see figure [ fig : directed - intuition ] ) , which implies two pairs are similar only if both pairs of endpoints are similar . this assumption should not to be confused with a different assumption made by many unsupervised link prediction methods , which assume that a link is more likely to exist between similar nodes , applicable to networks with assortative mixing .
assortative networks are common a typical example is a social network , where people commonly tend to be friends with those of similar age , income level , race , etc .
however , there are also networks with disassortative mixing , in which the assumption that similar pairs are more likely to be connected is no longer valid for example , predators do not typically feed on each other in a food web .
our assumption , in contrast , is equally plausible for both assortative and disassortative networks , as well as more general settings , as it does not assume anything about the relationship between @xmath18 and @xmath36 . motivated by this assumption of similar probabilities of links for similar node pairs , we propose to estimate @xmath37 by @xmath38 where @xmath39 is a real - valued @xmath40 matrix , and @xmath41 is a tuning parameter .
the first term is the usual squared error loss connecting the parameters with the observed network .
the minimizer of its population version , i.e. , @xmath42 is @xmath21 .
the second term enforces our key assumption , penalizing the difference between @xmath43 and @xmath44 more if two node pairs @xmath33 and @xmath34 are similar .
the choice of the squared error loss is not crucial , and other commonly used loss functions could be considered instead , for example , the hinge loss or the negative log - likelihood .
the main reason for choosing the squared error loss is computational efficiency , since it makes a quadratic problem ; see more on this details in section [ sec : alg ] . in some applications , we may have additional information about true positive and negative examples , i.e. , some @xmath45 s may be known to be true 1 s and true 0 s , while others may be uncertain .
this could happen , for example , when validation experiments have been conducted on a subset of a gene or protein network inferred from expression data .
if such information is available , it makes sense to use it , and we can then modify criterion as follows : @xmath46 where @xmath47 if it is known that @xmath48 , and 0 otherwise .
this is similar to a semi - supervised criterion proposed in @xcite .
however , @xcite did not consider the uncertainty in positive and negative examples , nor did they consider the undirected case which we discuss next . since only involves a partial sum of the loss function terms , we will refer to as the partial - sum criterion and as the full - sum criterion for the rest of the paper . for undirected networks , our key assumption that @xmath18 and @xmath31 are close if two pairs @xmath33 and @xmath34 are similar needs to take into account that the direction no longer matters ; thus the pairs are similar if either @xmath4 is similar to @xmath29 and @xmath5 is similar to @xmath32 , or if @xmath4 is similar to @xmath32 and @xmath5 is similar to @xmath29 ( see figure [ fig : undirected - intuition ] .
thus we need a new pair similarity measure that combines @xmath35 and @xmath49 .
there are multiple options ; for example , two natural combinations are @xmath50 empirically , we found that @xmath51 performs better than @xmath52 for a range of real and simulated networks .
the reason for this can be easily illustrated on the stochastic block model .
the stochastic block model is a commonly used model for networks with communities , where the probability of a link only depends on the community labels of its two endpoints .
specifically , given community labels @xmath53 , @xmath13 s are independent bernoulli random variables with @xmath54 where @xmath55 $ ] is a @xmath56 symmetric matrix , and @xmath57 is the number of communities in the network .
suppose we have the best similarity measure we can possibly hope to have based on the truth , @xmath58 , where @xmath59 is the indicator function .
in that case , implies @xmath60 if @xmath61 , whereas the sum of the weights would be misleading . using @xmath51 as the measure of pair similarity , we propose estimating @xmath26 for undirected networks by @xmath62 similarly to the directed case ,
if we have information about true positive and negative examples , we can use a partial - sum criterion @xmath63 where @xmath47 if it is known that @xmath48 , otherwise @xmath64 .
the last component we need to specify is the node similarity matrix @xmath30 .
one typical situation is when we have reasons to believe that the external node covariates are related to the structure of the network , in which case it is natural to use covariate information to construct @xmath65 .
though more complicated formats do exist , node covariates are typically represented by an @xmath66 matrix @xmath67 where @xmath68 is the value of variable @xmath69 on node @xmath4 .
then @xmath65 can be taken to be some similarity measure between the @xmath4-th and @xmath29-th rows of @xmath67 .
for example , if @xmath67 contains only numerical variables and has been standardized , we can use the exponential decay kernel , @xmath70 where @xmath71 is the euclidean vector norm .
when node covariates are not available , node similarity @xmath65 is usually obtained from the topology of the observed network @xmath11 , i.e. , @xmath65 is large if @xmath4 and @xmath29 have a similar pattern of connections with other nodes . for undirected networks ,
a simple choice of @xmath65 could be @xmath72 where @xmath73 denotes cardinality of a set .
this particular measure turns out to be not very useful : since most real networks are sparse , most entries of any @xmath69-th column will be 0 , and thus most of @xmath65 s would be large .
a more informative measure is the jaccard index @xcite , @xmath74 where @xmath75 is the set of neighbors of node @xmath4 .
the directed networks case is similar , except we need to count the in and the out links separately .
the formulas corresponding to and become @xmath76 where @xmath77 and @xmath78 .
the proposed link prediction criteria are convex and quadratic in parameters , and thus optimization is fairly straightforward .
the obvious approach is to treat the matrix @xmath39 as a long vector with @xmath79 elements ( or @xmath80 in the undirected case ) , and solve the linear system obtained by taking the first derivative of any criterion above with respect to this vector .
however , solving a system of linear equations could be challenging for large - scale problems @xcite ; the number of parameters here is @xmath81 , and so the linear system requires @xmath82 memory .
however , if @xmath30 is sparse , or sparsified by applying thresholding or some other similar method , then solving the linear system is the efficient choice .
if the @xmath30 matrix is not sparse , an iterative algorithm with sequential updates that only requires @xmath81 memory would be a better choice than solving the linear system .
we propose an iterative algorithm following the idea of block coordinate descent @xcite . a block coordinate descent algorithm partitions the coordinates into blocks and iteratively optimizes the criterion with respect to each block while holding the other blocks fixed .
first , we derive the update equations for directed networks .
note and can be written in the general form @xmath83 where @xmath84 for and @xmath85 for . for any matrix @xmath86 ,
let @xmath87 be the @xmath4th row of @xmath86 .
we treat @xmath88 as a block , and update @xmath89 iteratively . define @xmath90
. then @xmath91 let @xmath92 be an @xmath40 diagonal matrix with @xmath93 . then @xmath94 plugging and into , and taking the first derivative of @xmath95 with respect to @xmath89 , we obtain @xmath96 .
\nonumber\end{aligned}\ ] ] solving @xmath97 with respect to @xmath89 , we obtain the updating formula @xmath98 where @xmath99 is the value of @xmath100 at iteration @xmath0 . this update is fast to compute but its derivation relies on the product form of @xmath65 and @xmath101 , and thus is not directly applicable in the undirected case , where @xmath51 is used as the similarity measure .
however , we can still approximate @xmath51 with a product , using the fact that for @xmath102 , @xmath103 { x^q+y^q}= \max(x , y)$ ] .
thus , for sufficiently large @xmath104 , we have @xmath105^q \approx ( w_{ii'}w_{jj'})^q + ( w_{ij'}w_{ji'})^q . \label{appx}\end{aligned}\ ] ] further , @xmath106 is a monotone transformation of @xmath30 and can also serve as a similarity measure .
based on , we propose to substitute the following approximate criterion for undirected networks , @xmath107 where @xmath108 for the full sum criterion and @xmath85 for the partial sum criterion . by symmetry , @xmath109 this is now in the same form as , with each term in the sum containing a product of @xmath65 and @xmath101 , and therefore can be solved by block coordinate descent with an analogous updating equation as that in the directed network case . in practice
, we found that when @xmath30 is sparse or truncated to be sparse , solving the linear system can be much faster than the block coordinate descent method ; however , when @xmath30 is dense and the number of nodes is reasonably large , the block coordinate descent method dominates directly solving linear equations .
in this section , we test performance of our link prediction methods on simulated networks . in all cases ,
each network consists of @xmath110 nodes , and node @xmath4 s covariates @xmath111 are independently generated from a multivariate normal distribution @xmath112 with @xmath113 .
each @xmath114 is generated independently , with @xmath115 .
we consider the following functions @xmath116 : @xmath117 the right hand column gives sparser versions of functions in the left hand column ( subtracting a constant within the logit link functions lowers the overall degree ) , which we use to compare dense and sparse networks ( the average degrees of all these networks are reported in figures [ fig : sim1 ] and [ fig : sim2 ] ) .
functions ( a ) and ( b ) are asymmetric in @xmath111 and @xmath118 , giving directed networks , while ( c ) and ( d ) are symmetric functions corresponding to undirected networks .
further , @xmath119 and @xmath120 are linear functions ; @xmath121 is the projection model proposed in @xcite , under which the link probability is determined by the projection of @xmath111 onto the direction of @xmath118 , and @xmath122 is an undirected version of the projection model .
we also generate indicators @xmath123 s as independent bernoulli variables taking values 1 and 0 with equal probability , and set @xmath124 .
this setup corresponds to the `` partially observed '' network of the title , where all the observed edges are true but the missing edges may or may not be true 0s .
since we have node covariates affecting the probabilities of links in this case , we define the similarity matrix @xmath30 by @xmath70 where we choose @xmath125 .
after truncating @xmath30 at 0.1 , we optimize all criteria by solving linear equations , with @xmath41 chosen by 5-fold cross validation . the performance of link prediction is evaluated on the `` test '' set @xmath126 .
we report roc curves , which only depend on the rankings of the estimates @xmath127 rather than their numerical values .
specifically , let @xmath128 be the ranking of @xmath127 on the test set in descending order .
for any integer @xmath69 , we define false positives as pairs @xmath129 ranked within top @xmath69 but without links in the true network ( @xmath130 ) , and true positives as pairs ranked within top @xmath69 with @xmath131
. then the true positive rate ( tpr ) and the false positive rate ( fpr ) are defined by @xmath132 the roc curves showing the false positive rate vs. the true positive rate over a range of @xmath69 values are shown in figures [ fig : sim1 ] ( directed networks ) and [ fig : sim2 ] ( undirected networks ) .
each curve is the average of 20 replicates .
we also show the roc curve constructed from true @xmath18 s as a benchmark .. overall , both the full sum and the partial sum criteria perform well .
there is little difference between directed network models and their undirected versions .
as expected , the partial sum criterion always gives better results since it has more information and only uses the true positive and negative examples for training .
but its performance is quite comparable to the completely unsupervised full sum criterion , except perhaps for model @xmath120 .
the gaps between the unsupervised full sum criterion and semi - supervised partial sum criterion become smaller for sparse networks , as the false negatives in the full sum are only a small proportion of the large number of true negatives in a sparse network .
the roc curve obtained from the true model in sparse networks is better than in the corresponding dense networks ; this seemingly counter - intuitive finding is also explained by the large number of 0s in sparse networks .
however , gaps between both our link prediction methods and the true model are larger in all the sparse networks than in their dense counterparts .
this confirms the observation that a small number of positive examples in sparse networks makes the link prediction problem challenging .
our first application is to an undirected network containing yeast protein - protein interactions from @xcite .
this network was edited to contain only highly reliable interactions supported by multiple experiments @xcite , resulting in 984 protein nodes and 2438 edges , with the average node degree about 5 .
we take this verified network to be the true underlying network @xmath12 .
@xcite also constructed a matrix measuring similarities between proteins based on gene expression , protein localization , phylogenetic profiles and yeast two - hybrid data , which we use as the node similarity matrix @xmath30 for link prediction . here , we compare the full sum criterion , the partial sum criterion , and the latent variable model proposed by @xcite . to test prediction , we generate indicators @xmath123 s as independent bernoulli variables taking value 1 with probability @xmath16 , and set @xmath133 .
we consider three different values of @xmath16 , @xmath134 , corresponding to different amounts of available information .
we use the block coordinate descent algorithm proposed in section [ sec : alg ] to approximately optimize and , with @xmath135 and @xmath41 chosen by cross - validation .
the latent variable model depends on a tuning parameter @xmath57 , the dimension of the latent space .
we fix @xmath136 since larger values of @xmath57 do not significantly change the performance in this example .
we again use roc curves to evaluate the link prediction performance on the set @xmath126 .
each roc curve in figure [ fig : ppi ] is the average of 10 random realizations of @xmath123 s .
the semi - supervised criterion always performs better than the unsupervised criterion , as it should .
further , the semi - supervised criterion almost always outperforms the latent variable model , except for very small values of the false positive rate , and the fully unsupervised criterion also starts to outperform the latent variable model as the false positive rate increases .
the latent variable model is also more sensitive to the sampling rate @xmath16 , with performance deteriorating for @xmath137 .
this is because the model relies heavily on the structure of the network , and a low sampling rate may substantially distort the overall network topology .
on the other hand , we use the node similarity matrix @xmath30 which depends only on the features of the proteins , and is thus unaffected by the sampling rate .
this dataset is a school friendship network from the national longitudinal study of adolescent health ( see @xcite for detailed information ) .
this network contains 1011 high school students and 5459 directed links connecting students to their friends , as reported by the students themselves .
the average degree of this network is also around @xmath138 .
here we test our two link prediction criteria , with the same settings for @xmath123 as in the protein example .
since the latent variable model of @xcite is not applicable to directed networks , we omit it here . due to lack of node covariates
, we construct a network - based similarity @xmath30 by using the jaccard index defined in .
we again apply block coordinate descent to minimize the criteria with @xmath41 chosen by cross - validation , and report the average roc curves over 10 realizations of @xmath123 s . as shown in figure [ fig : school ] , both criteria perform fairly well for @xmath139 and @xmath140 , but fail for @xmath141 , as the sampling rate is too small for @xmath30 to capture the overall network topology .
this does not happen in the protein - protein interactions network , since @xmath30 is constructed from covariates on proteins and is unaffected by sub - sampling .
in this article , we have proposed a new framework for link prediction that allows uncertainty in observed links and non - links of a given network .
our method can provide relative rankings of potential links for pairs with and without observed links .
the proposed link prediction criteria are fully non - parametric and essentially model - free , relying only on the assumption that similar node pairs have similar link probabilities , which is valid for a wide range of network models .
one direction we would like to explore in the future is to combine more specific parametric network models with our non - parametric approach , with the goal of achieving both robustness and efficiency .
we are also investigating consistency properties of our method , which is challenging because it requires developing a novel theoretical framework for evaluating consistency of rankings .
we are also developing extensions that would allow the probabilities of errors , @xmath16 and @xmath17 , to depend on the underlying probabilities of links .
this would allow , for example , making highly probable links more likely to be observed correctly .
ultimately , we would also like to incorporate the general framework of link uncertainty into other network problems , for example , community detection .
h. kashima , t. kato , y. yamanishi , m. sugiyama , and k. tsuda .
link propagation : a fast semi - supervised learning algorithm for link prediction . in _ proceedings of the 2009 siam international conference on data mining _ , 2009 .
k. miller , t. griffiths , and m.i .
nonparametric latent feature models for link prediction . in y.
bengio , d. schuurmans , j. lafferty , and c. williams , editors , _ advances in neural information processing systems ( nips ) _ , volume 22 , 2010 .
k. yu , w. chu , s. yu , v. tresp , and z. xu . stochastic relational models for discriminative link prediction . in _ proceedings of neural information precessing systems _
, pages 15531560 . mit press , cambridge ma , 2007 . | link prediction is one of the fundamental problems in network analysis . in many applications , notably in genetics
, a partially observed network may not contain any negative examples of absent edges , which creates a difficulty for many existing supervised learning approaches .
we develop a new method which treats the observed network as a sample of the true network with different sampling rates for positive and negative examples .
we obtain a relative ranking of potential links by their probabilities , utilizing information on node covariates as well as on network topology .
empirically , the method performs well under many settings , including when the observed network is sparse .
we apply the method to a protein - protein interaction network and a school friendship network . |
the camassa - holm ( ch ) equation @xmath3 was discovered by camassa and holm @xcite as a model for unidirectional propagation of waves in shallow water , @xmath4 representing the height of the free surface about a flat bottom , @xmath5 being a constant related to the critical shallow water speed and @xmath6 a constant proportional to the mean water depth @xcite . equation ( [ ch ] ) had been previously found by fokas and fuchssteiner @xcite using the method of recursion operators and shown to be a bi - hamiltonian equation with an infinite number of conserved functionals .
it was also rediscovered by dai @xcite as a model for nonlinear waves in cylindrical hyperelastic rods , with @xmath4 representing the radial stretch relative to a pre - stressed state .
equation ( [ ch ] ) finally also arises in the study of the motion of a non - newtonian fluid of second grade in the limit when the viscosity tends to zero @xcite . a class of two
component generalizations of the ch equation has been recently obtained in @xcite . the initial value problem for ( [ ch ] ) @xmath7 presents interesting features : first for @xmath8 there may exist peakons , i.e. , non smooth solutions , second even for a smooth initial datum @xmath9 the wave - breaking phenomenon may occur ,
that is the solution @xmath4 remains bounded while its slope becomes unbounded in finite time .
this phenomenon was first noticed for @xmath8 by camassa and holm @xcite who showed that for smooth and odd initial datum @xmath9 such that @xmath10 for @xmath11 and @xmath12 , the slope @xmath13 is driven to @xmath14 in finite time . in the case @xmath8 , under the hypothesis that @xmath15 is smooth and summable , mckean @xcite proves that the wave - breaking phenomenon occurs if and only if some portion of the positive part of @xmath16 lies to the left of some portion of the negative part of @xmath16 .
the wave breaking phenomenon occurs also for @xmath17 and , in particular constantin @xcite shows that for initial data @xmath9 in the sobolev space @xmath18 , @xmath19 there exists a unique solution to ( [ ch ] ) @xmath20 defined for some maximal time @xmath21 ; moreover @xmath22 if and only if @xmath23 , i.e. , for @xmath24 , singularities in the solution may arise only in the form of wave breaking .
in this manuscript we are interested in studying the behaviour of the solution @xmath25 of the cauchy problem of the ch equation as @xmath26 for smooth initial data for which breaking does not occur . to this end
, we suppose that @xmath9 is in the schwartz class , has a single negative hump and satisfies the following non breaking condition @xmath27 in this case for any @xmath28 , @xmath29 is in the schwartz class and @xmath30 @xcite . for initial data in this class and @xmath31 , grava and klein @xcite
show that the numerical solution @xmath25 of ( [ ch ] ) develops a zone of fast oscillations , as for the small dispersion limit of the korteweg - de vries ( kdv ) equation , see fig .
[ chtime ] . in this paper , following the works of gurevich and pitaevskii @xcite , lax and levermore @xcite , deift , venakides and zhou @xcite , we claim and give numerical support for our claim that the description of the small dispersion limit of the ch equation follows 1)-3 ) below : \1 ) for @xmath32 , where @xmath33 is a critical time , the solution @xmath25 of the ch cauchy problem is approximated as @xmath34 , by @xmath4 which solves the hopf equation @xmath35 here @xmath33 is the time when the first point of gradient catastrophe appears in the solution to the hopf equation @xmath36 and it is given by the relation @xmath37}.\ ] ] \2 ) there exists @xmath38 such that for @xmath39 , the solution of the ch equation is characterized by the appearance of an interval of rapid oscillations .
if @xmath40 where @xmath41 is the inverse of the decreasing part of the initial datum @xmath9 and @xmath42 is the critical point , these oscillations are described in the following way . as @xmath26 , the interval of the oscillatory zone
is given by @xmath43 $ ] where @xmath44 are determined from the initial datum and satisfy the condition @xmath45 , with @xmath46 the coordinate of the point of the gradient catastrophe of the hopf solution .
in the ( x , t ) plane @xmath47 describe a cusp - shape region . outside the interval
@xmath43 $ ] the leading order asymptotics is given by the solution of the hopf equation .
inside the interval @xmath48 we claim that the solution @xmath25 is approximately described , for small @xmath0 by the one phase solution of ch which may be expressed in implicit form in terms of elliptic functions as @xmath49 with @xmath50 , @xmath51 and @xmath52 the complete jacobi elliptic integrals of the first and third kind of modulus @xmath53 , respectively , where @xmath54 the quantities @xmath55 , @xmath56 and the wave - number @xmath57 are abelian integrals defined respectively in ( [ abe : kappa ] ) , ( [ abe : p0 ] ) and ( [ abe : semi ] ) , the phase shift @xmath58 is defined in ( [ q ] ) .
@xmath59 is the third jacobi theta function of modulus @xmath60 ( see @xcite and ( [ vartheta3 ] ) ) . for constant values of the @xmath61 , ( [ abe : ellip ] )
is an exact solution to ch in implicit form ( see @xcite and section [ 11 ] ) .
moreover , for @xmath62 , such solution is real periodic and analytic in @xmath63 , so it is appropriate to call it the one phase solution .
however , unlike the kdv case , it can not be extended to meromorphic function on the @xmath63 complex plane .
indeed , the r.h.s of the first equation in ( [ abe : ellip ] ) is an elliptic function in @xmath64 , while @xmath65 is real analytic and invertible for real @xmath64 , but not meromorphic in @xmath66 . in the description of the leading order asymptotics of @xmath25 , as @xmath67 , the quantities @xmath61
depend on @xmath68 and evolve according to the ch modulation equations which were derived in @xcite for the one phase periodic solution . in terms of the riemann invariants @xmath69 they take the form @xmath70 with @xmath71 and @xmath72 . unlike the kdv case , the whitham equations for ch are not strictly hyperbolic and this fact gives some technical difficulties in their numerical and analytical treatment .
\3 ) near the left boundary of the cusp - shape region , in the double scaling limit @xmath73 and @xmath26 , in such a way that @xmath74 remains finite , the asymptotic solution of the ch equation is given by @xmath75 where @xmath4 is the solution of the hopf equation ( [ hopf ] ) , the phase @xmath76 is given in ( [ psi ] ) and the function @xmath77 is , up to shifts and rescalings , the hastings - mcleod solution to the painlev - ii equation @xcite @xmath78 determined uniquely by the boundary conditions @xmath79 with @xmath80 the airy function .
such a solution is real and pole free for real values of @xmath64 .
we verify numerically the validity of the above asymptotic expansions 1 ) , 2 ) and 3 ) for the initial datum @xmath81 and different values of @xmath5 , see for instance fig .
[ ch1e3asym ] .
+ * remark . * the asymptotic description of @xmath25 as @xmath26 near the critical time @xmath82 has been conjectured in @xcite and studied numerically in @xcite .
the asymptotic description of @xmath25 as @xmath26 at the right boundary of the cusp - shape region , namely near @xmath83 has not yet been studied even for the kdv case .
the paper is organized as follows : in the next section we compute the one phase solution to ch , in section [ 1 ] we obtain the small amplitude limit of the one phase solution to the ch equation . in section
[ 2 ] we perform a multiple scale analysis of the ch equations near the leading edge @xmath84 of the oscillatory zone . in section
[ 3 ] we present a numerical comparison of the small dispersion limit of the ch solution with the asymptotic formula ( [ abe : ellip ] ) .
a quantitative numerical comparison of the ch solution and the multiscale solution is presented in section [ 4 ] .
let us look for a one phase real periodic travelling wave solution to ( [ ch ] ) of the form @xmath85 where @xmath57 is the wave number , @xmath86 the frequency and @xmath87 is a phase to be determined from the initial conditions .
when we plug @xmath88 into the ch equation ( [ ch ] ) , after integration we get @xmath89 where @xmath90 and @xmath91 are constants of integration .
the ch one - phase solution @xmath92 is then obtained by inverting the differential of the third kind @xmath93 where @xmath94 .
let @xmath62 , so that @xmath88 is real periodic in the interval @xmath95 $ ] .
since @xmath96 has constant sign for @xmath97 $ ] , by a standard argument , the ( real ) inverse function @xmath88 exists and is monotone in @xmath98 $ ] , where @xmath99 is the half period of the travelling wave solution . to invert ( [ abe : solint ] ) and obtain @xmath100 let us introduce the elliptic curve @xmath101 with homology basis of cycles @xmath102 as in the figure below and let @xmath103 be the normalized holomorphic differential on @xmath104 , with @xmath51 the complete jacobi elliptic integral of modulus @xmath53 @xmath105 then inversion of the normalized holomorphic differential @xmath106 is given by the jacobi theorem @xcite and takes the form @xmath107 where @xmath108 @xmath109 @xmath110 is the third jacobi - theta function defined by the fourier series @xmath111,\quad\quad \tau = i \frac{k^{\prime}(s)}{k(s)},\ ] ] @xmath112 and @xmath113 is defined by @xmath114 the function @xmath115 is an elliptic function in @xmath64 with periods @xmath116 and @xmath117 and it is real periodic in @xmath64 along the complex line @xmath118 , @xmath119 . finally , inserting ( [ abe : jacobi ] ) into ( [ abe : solint ] ) , we get @xmath120 equations ( [ abe : jacobi ] ) and ( [ solution ] ) give the one phase solution @xmath121 in implicit form .
the real half period , @xmath99 , of @xmath121 is @xmath122 normalizing the half period @xmath99 to @xmath123 , from ( [ abe : semi ] ) we obtain the wave number and the frequency @xmath124 summarizing , the one phase real nonsingular solution to the ch equation takes the form @xmath125 observe from the above formula that @xmath25 is not a meromorphic function of @xmath126 and @xmath127 . an alternative method to obtain @xmath25
is discussed in @xcite where the solution is expressed in terms of conveniently generalized theta - functions in two variables , which are constrained to the generalized theta - divisor of a 2dimensional generalized jacobian .
the envelope of the oscillations is obtained by the maximum and minimum values of the theta function and gives @xmath128
we analyze the one - phase solution ( [ abe : ellisol ] ) near the leading edge @xmath84 , namely when the oscillations go to zero . for this purpose
we need to study the ch modulation equations more in detail . in @xcite ,
abenda and grava constructed the one - phase ch modulation equations and showed that @xmath61 , @xmath129 are the riemann invariants .
the ch modulation equations take the riemann invariant form @xmath130 where the speeds @xmath131 in ( [ abe : whi ] ) are explicitly given by the formula @xcite @xmath132 } , \end{split}\ ] ] with @xmath71 , @xmath51 , @xmath52 as before and @xmath133 the complete elliptic integral of the second kind with modulus @xmath134 . in the limit when two
riemann invariants coalesce , the modulation equations reduce to the hopf equation @xmath135 the ch modulation equations are integrable via the generalized hodograph transform introduced by tsarev @xcite , @xmath136 which gives the solution of ( [ whithamu ] ) in implicit form .
the formula of the @xmath137 for the specific ch case , has been obtained in @xcite , @xcite : @xmath138 where the function @xmath139 is given by @xmath140 with @xmath41 the inverse of the decreasing part of the initial datum @xmath9 .
the above formula of @xmath141 is valid for @xmath142 where @xmath143 is the coordinate of the minimum value of the solution @xmath4 of the hopf equation .
for @xmath144 the corresponding formula for @xmath141 contains also the increasing part of the initial datum ( see @xcite ) . in @xcite
it is shown that if @xmath145 , then the solution @xmath146 of the system ( [ abe : wi ] ) exists for some time @xmath147 . in order to take the small amplitude limit of the solution ( [ abe : ellip ] ) , we rewrite the system ( [ abe : hodosol ] ) in the equivalent form @xcite @xmath148 where @xmath149 with @xmath53 and @xmath150 as in ( [ abe : modulus ] ) and ( [ elli3 ] ) , and we perform the limit @xmath151 , where @xmath152 let @xmath153 then the following identities hold @xmath154 substituting ( [ abe : quv ] ) , ( [ ident ] ) and the expansion of the elliptic integrals @xmath133 , @xmath51 and @xmath52 as @xmath155 @xcite into ( [ abe : hodonodeg ] ) , we arrive to the system @xmath156 from the above , we deduce that , in the limit @xmath151 , the hodograph transform ( [ abe : hodosol ] ) reduces to the form @xmath157 the above system of equations enables one to determine @xmath126 , @xmath158 and @xmath159 as functions of time .
we denote this time dependence as @xmath160 , @xmath161 and @xmath162 .
observe that @xmath163 where @xmath4 is the solution of the hopf equation . in what follows
we will always denote by @xmath158 or @xmath164 the solution of the hopf equation @xmath4 at the leading edge @xmath160 while we will refer to the solution of the ch equation as @xmath25 .
the derivatives with respect to time of these quantities are given by @xmath165 with @xmath166 we are interested in studying the behaviour of the one phase solution ( [ abe : ellisol ] ) near the leading edge , namely when @xmath167 . to this aim ,
we introduce two unknown functions of @xmath126 and @xmath127 , @xmath168 which tend to zero as @xmath169 .
we now derive the dependence of @xmath170 as a function of @xmath171 .
let us fix @xmath172 for the first equation in ( [ abe : hodonodeg ] ) near @xmath173 , we find @xmath174 we then substitute @xmath175 in the above equation , we insert ( [ abe : quv ] ) , ( [ ident ] ) and ( [ delta0 ] ) into it , and we get @xmath176 so that @xmath177 similarly , using the second equation in ( [ abe : hodonodeg ] ) , we arrive at @xmath178 with @xmath179 defined in ( [ v ] ) . in the limit @xmath180 the one - phase solution of the ch equation , implicitly defined by ( [ abe : ellip ] ) , has the following trigonometric expansion @xmath181 with @xmath182 _ proof : _ we first prove ( [ zeta ] ) .
let @xmath183 in ( [ abe : ellip ] ) , so that @xmath184,\ ] ] with @xmath57 as is ( [ abe : kappa ] ) , @xmath141 as in ( [ q ] ) . in the limit @xmath185 , with @xmath186 , @xmath187 , @xmath188
, we get @xmath189 where we use @xmath190 to estimate the error in the formula above . then
, we insert the solution @xmath191 of equations ( [ delta0 ] ) , we use ( [ deltax ] ) and ( [ deltax ] ) to estimate @xmath170 and @xmath192 and we get @xmath193 where @xmath194 is defined in ( [ abe : quv ] ) , which coincides with ( [ zeta ] ) .
we now prove that in the limit @xmath151 , @xmath25 , defined implicitly as a function of @xmath63 in ( [ abe : ellip ] ) , has the following trigonometric series expansion @xmath195 to this aim , we expand @xmath25 and @xmath63 defined in ( [ abe : ellip ] ) in trigonometric series of @xmath64 and we compute @xmath64 as a function of @xmath63 .
we need the following series expansions for the jacobi theta functions @xcite @xmath196 where @xmath56 is as in ( [ abe : p0 ] ) and @xmath197 inserting the above expansion and @xmath198 into ( [ abe : thetaexp ] ) , we get @xmath199 from which we find @xmath200 and the assertion easily follows .
in this section we propose a multiscale description of the oscillatory behavior of the solution to the ch equation in the small dispersion limit ( @xmath67 ) close to the leading edge @xmath191 where @xmath202 and @xmath203 .
we follow closely the approach @xcite for the corresponding kdv situation .
the ansatz for the multiscale expansion to the ch solution close to the leading edge is inspired by the asymptotic solution in the whitham zone discussed in the previous section .
numerically we find that the quantity @xmath204 in ( [ ellilim ] ) is of the order @xmath2 .
this implies with ( [ deltax ] ) that @xmath205 in the double scaling limit @xmath67 and @xmath206 .
we are thus led to introduce the rescaled coordinate @xmath207 near the leading edge , @xmath208 which transforms the ch equation ( [ ch ] ) to the form @xmath209 where @xmath210 numerically the corrections to the hopf solution near the leading edge are of order @xmath2 and thus we make as in @xcite the ansatz @xmath211 where @xmath212 is the solution of the hopf equation at the leading edge .
we assume that the terms @xmath213 , @xmath214 contain oscillatory terms of the order @xmath215 and take the form @xmath216 where @xmath217 terms proportional to @xmath218 in @xmath219 can be absorbed by a redefinition of @xmath76 , the higher order terms are chosen to compensate terms of lower order appearing in the solution to ch due to the non - linearities .
we immediately find @xmath220 from the terms of order @xmath221 and @xmath222 .
the term of order @xmath223 gives @xmath224 in particular , ( [ eq13 ] ) is algebraic in @xmath225 with coefficients depending on @xmath127 only , i.e. , @xmath226 , with @xmath227 to be determined . at the order @xmath2 , we get @xmath228 and @xmath229 @xmath230 at the order @xmath231 , using ( [ eq13 ] ) to ( [ eq15 ] ) , we find @xmath232 and @xmath233 @xmath234 with @xmath235 integration constant , @xmath236 we determine @xmath225 inserting ( [ ut ] ) inside ( [ eq15 ] ) , so that @xmath237 where the sign in front of the square root has to be chosen in such a way that the r.h.s . is positive
. then comparing ( [ psi2pm ] ) , ( [ psi ] ) , ( [ zeta ] ) and ( [ y ] ) , we conclude that @xmath238 inserting ( [ psi2 ] ) into ( [ eq13 ] ) , we find @xmath239 by comparison with ( [ zeta ] ) , we observe that @xmath240 then integrating the l.h.s . of the above expression between @xmath33 and @xmath127
we arrive at ( [ zeta ] ) , namely @xmath241 moreover , consistency between ( [ chexp ] ) and ( [ ellilim ] ) implies that @xmath242 from which we immediately verify that @xmath243 @xmath244 with @xmath245 in ( [ b1 ] ) .
summarizing , we get @xmath246+o(\epsilon),\ ] ] where @xmath247 with @xmath248 defined in ( [ abe : quv ] ) and @xmath249 satisfies the painlev - ii equation @xmath250 with @xmath251 using ( [ ut ] ) , we get @xmath252 with @xmath179 defined in ( [ v ] ) . making the substitution @xmath253 , @xmath254 with @xmath255 we arrive to the special painlev ii equation in normal form @xmath256 for @xmath257 , from the small amplitude limit of the one phase solution to the ch equation we get @xmath258 with @xmath259 defined in ( [ deltax ] ) , so that in the limit @xmath260 ( equivalently @xmath261 or @xmath262 ) , we have @xmath263 if @xmath264
, the ch solution is approximately the solution to the hopf equation so that @xmath265 , for @xmath266 ( equivalently @xmath267 )
. therefore we conclude that @xmath268 the solution to the painlev ii equation satisfying such asymptotic condition exists and is unique and pole free on the real line according to hastings and mcleod @xcite .
in this section we will solve numerically the ch and the whitham equations for initial data in the schwartzian class of rapidly decreasing functions with a single negative hump . as a concrete example we will study the initial datum @xmath269 for this initial datum the non
breaking condition ( [ nonbreak ] ) is satisfied if @xmath270 .
the resolution of the rapid modulated oscillations in the region of a dispersive shock is numerically demanding .
the strong gradients in the oscillatory regions require efficient approximation schemes which do not introduce an artificial numerical dissipation into the system .
we therefore use fourier spectral methods which are known for their excellent approximation properties for smooth functions whilst minimizing the introduction of numerical viscosity .
we restrict to initial data , where @xmath271 does not change sign , to ensure analyticity of the ch solutions .
the schwartzian solutions can be treated as effectively periodic if the computational domain is taken large enough that the solution is of the order of machine precision ( @xmath272 in matlab ) at the boundaries .
we always choose the computational domain in this way to avoid gibbs phenomena at the boundaries .
even with spectral methods , a large number of fourier modes is needed to resolve the rapid oscillations numerically . to obtain also a high resolution in time ,
we use high order finite difference methods , here a fourth order runge - kutta scheme . for stability reasons
a sufficiently small time step has to be chosen .
unconditionally stable implicit schemes could be used instead , but these can be only of second order .
such approaches would be too inefficient for the precision requested .
notice that the terms with the highest derivative in ch ( [ ch ] ) are not linear in contrast to the kdv equation ( we choose the kdv equation in a way that it has the same dispersionless equation as ( [ ch ] ) , the term proportional to @xmath5 can be always eliminated here in contrast to ch by a galilean transformation ) @xmath273 for such equations efficient integration schemes are known .
in @xcite it was shown that _ exponential time differentiation _ ( etd ) methods @xcite are the most efficient in the small dispersion limit of kdv .
for ch such an approach is not possible due to the nonlinearity of the highest order derivatives .
but we can use analytic knowledge of the solution : the dispersionless equation , the hopf equation , will have a point of gradient catastrophe at the critical time @xmath274 , for the example ( [ initial ] ) @xmath275 . for times
@xmath276 , the ch solution is very close to the hopf solution with moderate gradients . in this case
we can use a larger time step . for times close to the critical time and beyond
( in our example @xmath277 ) , we have to use considerably smaller time steps .
the main difference to the kdv equation is the non - locality of ch , which provides some filtering for the high frequencies : if we denote the fourier transform of @xmath158 with respect to @xmath126 by @xmath278 , we get for ( [ ch ] ) in fourier space @xmath279 the term @xmath280 in the above equation is the reason why ch gives for large spatial frequencies a better approximation to one - dimensional wave phenomena than kdv with a term proportional to @xmath281 in fourier space .
its effect in the present context is twofold : first it provides a high frequency filtering which allows in practice for larger time steps in the computation .
secondly it suppresses the rapid modulated oscillations in the shock region of the dispersionless equation .
this can be seen in fig .
[ kdvch ] , where solutions to kdv and ch for the same initial data are shown .
the kdv solutions always show much more oscillations than the ch solutions .
notice that despite the lower number of oscillations in the ch solutions in fig .
[ kdvch ] , depending on the value of @xmath5 a higher number of fourier modes is needed to resolve the oscillations .
this is due to the fact that the ch solution is less smooth than the kdv solution and thus less localized in fourier space , as can be seen in fig .
[ kdvch2 ] . therefore , to treat the same values of the small dispersion parameter @xmath0 for ch as for kdv , we need higher temporal and spatial resolution .
it is thus computationally more demanding to obtain the same number of oscillations in ch solutions as in kdv solutions , the latter being important to obtain a valid statistics for the scaling studied below .
the quality of the numerics is controlled via energy conservation for ch , @xmath282 the numerically computed energy will depend on time due to numerical errors .
as was discussed in @xcite , energy conservation can thus be used to check numerical accuracy . in practice energy conservation
overestimates numerical precision by 1 - 2 order of magnitude .
we typically solve the ch equation with a relative numerical error @xmath283 .
this ensures that the difference between the numerical and the asymptotic ch solution , which is typically of the order of @xmath0 or larger , is entirely due to the asymptotic description .
the whitham equations ( [ abe : whi ] ) have a similar form as the respective equations for kdv .
we use the same procedure to solve them numerically : we first solve the equations at the edges of the whitham zone and then at intermediate points .
for details the reader is referred to @xcite .
typical solutions for the whitham equations can be seen in fig .
[ whithb ] . in the shown example
, the quantity @xmath284 crosses the hump at @xmath285 .
in contrast to kdv , the system ( [ abe : whi ] ) is not strictly hyperbolic , i.e. , the speeds @xmath286 , @xmath129 , in ( [ ci1 ] ) do not satisfy for all @xmath126 and @xmath287 the relation @xmath288 .
in fact the lines @xmath289 can cross for a given time for @xmath290 as can be seen in fig .
[ chspeeds ] .
for @xmath291 the speeds @xmath292 and @xmath293 intersect in the shown example in the interior of the whitham zone as can be seen in more detail in fig .
[ chspeedsm ] .
this behavior of the speeds has no influence on the numerical solubility of the whitham equations .
it also does not influence the quality of the asymptotic solution in these cases as can be seen in fig .
[ chasym2nu ] .
the asymptotic description of the small dispersion limit of the ch equation is as follows : for times @xmath294 , the hopf solution for the same initial datum provides an asymptotic description . for @xmath287 ,
the whitham zone opens . outside this zone
, the hopf solution again serves as an asymptotic solution . in the interior
the one - phase solution to the ch equation describes the oscillatory behavior .
it is given on an elliptic curve with branch points being solutions of the whitham equations .
below we will study the validity of this asymptotic description in various regions of the ( @xmath68)-plane . to study the @xmath0 dependence of a certain quantity @xmath90 , we perform a linear regression analysis for the dependence of the logarithms , @xmath295 .
we compute all studied quantities for the @xmath0 values @xmath296 with @xmath297 $ ] .
generally it is found that the correlations and the standard deviations are worse than in the kdv case due to the lower number of oscillations . _ before breakup , _
@xmath298 + for times much smaller than the critical time , we find that the @xmath299 norm of the difference between hopf and ch solution decreases as @xmath300 .
more precisely we find by linear regression an exponent @xmath301 with correlation coefficient @xmath302 and standard deviation @xmath303 .
_ at breakup , _
@xmath304 + for times close to the breakup time , the hopf solution develops a gradient catastrophe .
the largest difference between hopf and ch solution can be found close to the breakup point .
we determine the scaling of the @xmath299 norm of the difference between hopf and ch solution on the whole interval of computation .
we find that its scaling is compatible with @xmath305 as conjectured in @xcite .
more precisely we find in a linear regression analysis @xmath306 ( @xmath307 ) with a correlation coefficient @xmath308 and standard deviation @xmath309 . an enhanced asymptotic description of the ch solution near the breakup point in terms of a solution to the painlev i2 equation was conjectured in @xcite and studied numerically in @xcite .
_ times _ @xmath287 : + for times @xmath310 it can be seen from fig .
[ ch1e3asym ] and fig .
[ chasym2nu ] that the asymptotic solution gives a very satisfactory description of the oscillations except at the boundaries of the whitham zone .
the asymptotic solution is so close to the approached solution that one can only see discrepancies near the boundary of the whitham zone , where the asymptotic solution is just @xmath311 .
thus one has to consider the difference between the solutions as shown in fig .
[ chdelta4e ] .
the quality of the numerics allows the study of the scaling behavior at various points in the whitham zone .
_ interior of the whitham zone : _
+ if we study the @xmath0-dependence of the @xmath299-norm of the difference near the middle of the whitham zone ( we take the maximum of this difference near the geometric midpoint ) , we find that the norm scales as @xmath0 .
more precisely we find an exponent @xmath312 with correlation coefficient @xmath313 and standard deviation @xmath314 .
there is obviously a certain arbitrariness in our definition of this error especially for larger values of @xmath0 , where there are only few oscillations . in these cases
the errors are read off close to the boundaries of the whitham zone where much smaller exponents for the error are observed ( see below ) .
this scaling gives nonetheless strong support for the conjectured form of the phase of the one - phase solution in the whitham zone , since even a small analytical error in the phase would lead to large errors which would not decrease with @xmath0 .
_ leading edge of the whitham zone : _
+ oscillations can always be found outside the whitham zone , whereas the hopf solution does not show any oscillations .
the biggest difference always occurs at the boundary of the whitham zone .
it scales as @xmath2 .
more precisely we find in the hopf zone @xmath315 with @xmath316 and @xmath314 . in the interior of the whitham zone
, @xmath249 has a similar value , but the correlation is worse . if one studies the scaling of the zone , where the difference between hopf and ch solution near the leading edge has absolute value larger than some treshold ( we use @xmath317 ) , we find a decrease compatible with @xmath231 , more precisely @xmath318 with @xmath316 and @xmath319 . _
trailing edge of the whitham zone : _ + the biggest difference is always found at the boundary of the whitham zone .
its scaling is compatible with @xmath320 .
we find @xmath321 with @xmath322 and @xmath323 .
in this section we study numerically the multiscales solution to ch derived in section [ 2 ] .
it will be shown that the latter provides a better description of the asymptotic behavior near the leading edge of the whitham zone than the hopf or the one - phase solution to ch as discussed in section [ 3 ] .
we identify the zone , where the latter gives a better description of ch than the former and study the @xmath0-dependence of the errors . for the numerical computation of the hastings - mcleod solution and a comparison to the kdv case
, we refer the reader to @xcite . in fig .
[ ch3in1 ] we show the ch solution , the asymptotic solution via whitham and hopf and the multiscales solution near the leading edge of the whitham zone .
it can be seen that the one - phase solution gives a very good description in the interior of the whitham zone as discussed in section [ 3 ] , whereas the multiscales solution gives as expected a better description near the leading edge . in fig .
[ chp21e4 ] the ch and the multiscales solution are shown in one plot for @xmath324 .
it can be seen that the agreement near the edge of the whitham zone is so good that one has to study the difference of the solutions .
the solution only gives locally an asymptotic description and is quickly out of phase for larger distances from the leading edge .
the difference between ch and multiscales solution is shown for several values of @xmath0 in fig .
[ chp21e4diff4e ] .
it can be seen that the maximal error still occurs close to the whitham edge , but that it decreases much faster with @xmath0 than the error given by the hopf and the whitham solution . a linear regression analysis for the logarithm of the difference @xmath170 between ch and multiscales solution near the edge
gives a scaling of the form @xmath325 with @xmath326 with standard deviation @xmath327 .
since there are much less oscillations in the ch case than in kdv , the found statistics is considerably worse in the former case than in the latter @xcite , which is reflected by the low correlation coefficient @xmath328 and the comparatively large standard deviation .
nonetheless the found scaling is in accordance with the @xmath231 scaling expected from the multiscales expansion . as can be already seen from fig .
[ ch3in1 ] , the multiscales solution gives a better asymptotic description of ch near the leading edge of the whitham zone than the hopf and the one phase solution .
this is even more obvious in fig .
[ ch1e42d ] where the difference between ch and the asymptotic solutions is shown .
this suggests to identify the regions where each of the asymptotic solutions gives a better description of ch than the other .
the results of this analysis can be seen in fig .
[ ch1e4match ] .
this matching procedure clearly improves the ch description near the leading edge . in fig .
[ chmatchp2 ] we see the difference between this matched asymptotic solution and the ch solution for two values of @xmath0 .
visibly the zone , where the solutions are matched , decreases with @xmath0 .
there is a certain ambiguity in the precise definition of this matching zone due to the oscillatory character of the solutions . because of the much lower number of oscillations than in kdv , the statistics is considerably worse in the ch case than in the kdv case .
the limits of the matching zone for several values of @xmath0 can be seen in fig .
[ width ] . due to the lower number of oscillations in the hopf region
, the matching zone extends much further into this region than in the whitham region .
the width of this zone scales like @xmath329 with @xmath321 and standard deviation @xmath303 and correlation coefficient @xmath316 .
this work has been supported by the misgam program of the european science foundation and the project from - pde funded by the european research council through the advanced investigator grant scheme .
ck thanks for financial support by the conseil rgional de bourgogne via the faber scheme . ,
_ ` numerical study of a multiscale expansion of kdv and camassa - holm equation ' _ , in integrable systems and random matrices , ed . by j. baik , t. kriecherbauer , l .- c .
li , k.d.t - r . mclaughlin and c. tomei , contemp
458,(2008 ) pp . | the small dispersion limit of solutions to the camassa - holm ( ch ) equation is characterized by the appearance of a zone of rapid modulated oscillations . an asymptotic description of these oscillations is given , for short times , by the one - phase solution to the ch equation , where the branch points of the corresponding elliptic curve depend on the physical coordinates via the whitham equations .
we present a conjecture for the phase of the asymptotic solution .
a numerical study of this limit for smooth hump - like initial data provides strong evidence for the validity of this conjecture .
we present a quantitative numerical comparison between the ch and the asymptotic solution .
the dependence on the small dispersion parameter @xmath0 is studied in the interior and at the boundaries of the whitham zone . in the interior of the zone ,
the difference between ch and asymptotic solution is of the order @xmath0 , at the trailing edge of the order @xmath1 and at the leading edge of the order @xmath2 . for the latter we present a multiscale expansion which describes the amplitude of the oscillations in terms of the hastings - mcleod solution of the painlev ii equation .
we show numerically that this multiscale solution provides an enhanced asymptotic description near the leading edge .
small dispersion limit , whitham equations , painlev transcendents , multiple scale analysis primary , 65m70 ; secondary , 65l05 , 65m20 |
proposition 4.15 of the monograph _ knots _ @xcite erroneously asserts and proves that a satellite knot is fibered if and only if it has nonzero winding number and both its companion and pattern knots are fibered . in 1990 , the third author of the present paper contacted the authors of _ knots _ , informed them of the mistake , and provided a counterexample .
the recent second edition of the now - classic textbook acknowledged the error and reproduced the example .
however , the necessary and sufficient conditions that it provided are not the most convenient in practice .
the present paper is an expanded version of an unpublished note that the third author wrote in 1991 .
its algebraic necessary and sufficient conditions for a satellite knot to fiber , do not appear to be well known .
( it is briefly mentioned on pages 221222 of @xcite . )
theorem 4.2 of the monograph @xcite by d. eisenbud and w. neumann gives topological necessary and sufficient conditions for a satellite knot in a homology @xmath3-sphere to fiber .
stated in terms of splice components " arising from the jaco - shalen johannson splitting theorem , it requires that each component be fibered and that the fibers meet each other compatibly .
an alternative , topological condition presented here ( see theorem [ alternate ] ) is more convenient in many situations .
it was discovered by the first author , who proved it using d. gabai s sutured manifold theory .
an elementary group theoretic proof was subsequently found by the second author .
we work throughout in the smooth category . as usual
, a _ knot _ is an embedded circle in the @xmath3-sphere @xmath4 .
it is _ trivial _ if it is the boundary of an embedded disk in @xmath4 .
more generally , a _ link _ is a union of finitely many pairwise disjoint knots .
when each component is provided with a direction , the link is said to be _
oriented_. the complement of an open tubular neighborhood of a knot or link @xmath5 is called its _ exterior _ , and it is denoted by @xmath6 .
its boundary @xmath7 can be identified with @xmath8 .
a knot @xmath9 is _ fibered _ if the projection @xmath10 extends to a fibration of @xmath11 .
in such a case , @xmath11 is the total space of a fiber bundle over the circle @xmath12 . by theorems of l. neuwirth @xcite and j. stallings @xcite ,
@xmath9 is fibered if and only if the commutator subgroup @xmath13 $ ] of the knot group @xmath14 is finitely generated and hence free .
the commutator subgroup is the kernel of the abelianization homomorphism @xmath15 sending the class of an oriented meridian of @xmath9 to @xmath16 .
satellite knots were introduced by h. schubert @xcite as a generalization of knot product .
we briefly review their definition and some key properties .
let @xmath0 be a knot embedded in a standard solid torus @xmath17 .
assume that @xmath0 is not isotopic to @xmath18 nor is it contained in any @xmath3-ball of @xmath1 .
let @xmath19 be a homeomorphism from @xmath1 onto a closed tubular neighborhood of a nontrivial knot @xmath20 , mapping a longitude of @xmath1 onto a longitude of @xmath20 .
( by a _ longitude _ of @xmath1 , we mean a meridian of the complementary solid torus @xmath21 . )
the image @xmath22 is called a _ satellite knot _ with _ companion knot _ @xmath20 and _ pattern _ @xmath2 . by a _ pattern knot
_ we mean @xmath0 regarded in @xmath23 .
the _ winding number _ of @xmath0 in @xmath1 is the nonnegative integer @xmath24 such that the homomorphism @xmath25 induced by inclusion has image @xmath26 .
two patterns @xmath27 and @xmath28 are regarded as the same if @xmath29 and @xmath30 are isotopic in @xmath1 .
we denote the various groups @xmath31 and @xmath32 by @xmath33 and @xmath34 , respectively . by the seifert - van kampen theorem , @xmath35 the amalgamated free product of @xmath34 and @xmath36 with subgroups @xmath37 and @xmath38 identified via @xmath19 .
it is convenient to regard @xmath34 as the group of the @xmath39-component link @xmath40 , where @xmath41 is a meridian of @xmath1 .
let @xmath42 be a longitude of @xmath1 , and let @xmath43 denote its homotopy class in @xmath34 .
the quotient group @xmath44 is naturally isomorphic to @xmath45 .
we define @xmath46 to be the kernel of @xmath47 , where @xmath48 and @xmath49 are the quotient and abelianization homomorphisms , respectively . when the winding number @xmath24 is nonzero , then the reidemeister - schreier method can be used to show that the commutator subgroup @xmath50 decomposes as an amalgamated free product : @xmath51 here @xmath52 is the class of a meridian of @xmath9 , while @xmath53 is the subgroup of @xmath46 generated by @xmath54 .
( for a proof , see section 4.12 of @xcite . )
let @xmath55 denote the complement of an open tubular neighborhood of @xmath0 in @xmath1 .
we say that the pattern @xmath2 is _ fibered _ if there exists a fibration @xmath56 inducing the homomorphism @xmath47 on fundamental groups .
it is well known ( and not difficult to see ) that in this case , the winding number of @xmath0 in @xmath1 is nonzero .
moreover , @xmath46 is the fundamental group of the fiber , a compact surface with boundary , and hence @xmath46 is free .
the following fibering criterion for satellite knots is implicit in the knot theory literature ( e.g. @xcite ) , but to our knowledge it has not previously appeared .
[ criterion1 ] let @xmath9 be a satellite knot with companion knot @xmath20 and pattern @xmath2 .
then the following conditions are equivalent . 1
. @xmath9 is fibered ; 2 .
@xmath57 and @xmath46 are finitely generated ; 3 .
@xmath20 and @xmath2 are fibered .
_ suppose that @xmath9 is fibered .
the winding number @xmath24 of @xmath0 in @xmath1 must be nonzero .
for otherwise , @xmath58 , which is a subgroup of @xmath59 by equation [ g ] , would be contained in the free commutator subgroup @xmath50 an impossibility .
consequently , @xmath50 has the form described by equation [ g ] .
since a free product of groups amalgamated over a finitely generated group is itself finitely generated if and only if each of its factors is ( see lemma 4.14 of @xcite , for example ) , both @xmath57 and @xmath46 are finitely generated .
if @xmath57 and @xmath46 are finitely generated , then stallings s theorem @xcite implies that @xmath20 and @xmath2 are fibered .
finally , if @xmath20 and @xmath2 are fibered , then @xmath57 and @xmath46 are isomorphic to fundamental groups of fibers for respective fibrations .
consequently , both groups are finitely generated .
equation [ g ] implies that the commutator subgroup @xmath50 is finitely generated . by stallings
s theorem @xcite , the satellite knot @xmath9 is fibered .
let @xmath60 be an oriented link .
a meridian of any component acquires an orientation by the right - hand rule .
the link is _ fibered _ if its exterior @xmath6 admits a fibration @xmath61 .
such a fibration induces an epimorphism @xmath62 , and as usual we require that the class of each meridian be sent to @xmath16 .
the epimorphism @xmath62 is the abelianization of an epimorphism @xmath63 .
by @xcite and @xcite , the link @xmath5 is fibered if and only if the kernel @xmath64 of @xmath65 is finitely generated .
the group @xmath64 is known as the _ augmentation subgroup _ of the group of @xmath5 . [
oriented ] orientation is not required in the definition of a fibered knot , since any knot is fibered if and only if the knot with reversed orientation is fibered .
similarly , an orientated link is fibered if and only if the link obtained by reversing the orientation of every component is fibered .
however , reversing the orientation of some but not all components can destroy the condition of fiberedness .
( a simple example is provided by the closure of the @xmath39-braid @xmath66 . )
consider a pattern @xmath67 such that @xmath0 has nonzero winding number @xmath24 .
we associate an oriented @xmath3-component link @xmath68 , where @xmath41 and @xmath69 are disjoint meridians of @xmath1 with opposite orientations .
the orientation of @xmath0 is arbitrary .
[ alternate ] the pattern @xmath2 is fibered if and only if @xmath70 is a fibered link .
_ as in section 2 , let @xmath34 be the group of the @xmath39-component link @xmath40 , and let @xmath46 be the kernel of the epimorphism @xmath71 mapping the class of a meridian of @xmath0 to @xmath16 and @xmath43 , the class of a meridian of @xmath41 , to @xmath72 .
it suffices to show that the augmentation subgroup @xmath64 of the link @xmath73 is the free product of @xmath46 and a finitely generated group ( in fact , a free group of rank @xmath24 , the winding number of @xmath0 in @xmath1 ) .
since the free product of two groups is finitely generated if and only if each of its factors is finitely generated , it will then follow that @xmath46 is finitely generated if and only if @xmath64 is .
equivalently , the pattern @xmath2 is fibered if and only if the link @xmath70 is fibered .
part of a diagram @xmath74 for @xmath75 appears in fig .
[ d2 ] ( a ) with families of generators indicated .
there are @xmath76 strands of @xmath0 passing the meridional disk bounded by the unknotted circle representing @xmath41 .
assign a weight @xmath77 of @xmath16 or @xmath72 to each arc of @xmath74 according to whether the arc belongs to @xmath0 or @xmath41 , respectively .
the wirtinger algorithm combined with the reidemeister - schreier method enables us to write a presentation for @xmath46 .
each arc represents an infinite family @xmath78 @xmath79 of generators , indexed by @xmath80 .
the letters @xmath81 correspond to arcs of @xmath0 , with one arc labeled only with the identity element ; the remaining letters correspond to arcs of @xmath41 .
relators , which come in families indexed by @xmath82 , correspond to the crossings of @xmath74 , and have the form @xmath83 , where @xmath77 and @xmath84 are the weights of the arcs corresponding to @xmath85 and @xmath86 , respectively .
we use relators @xmath87 ( indicated in the figure by boxed numbers ) to eliminate @xmath79 .
recall that in a link diagram , any one wirtinger relator is a consequence of the others . consequently , any single relator family is a consequence of the others .
we regard @xmath88 as a redundant family of relators .
the relator families arising from the remaining crossings in the figure have the form @xmath89 , where @xmath90 denotes inverse .
consider now the partial diagram @xmath91 for @xmath92 that appears in fig .
[ d2 ] ( b ) , again with families of generators indicated .
letters @xmath93 correspond to arcs of @xmath0 while @xmath94 and @xmath95 correspond to arcs of @xmath41 and @xmath69 , respectively .
the parts of @xmath74 and @xmath91 not shown are identical .
, ( b ) partial diagram of @xmath91.,height=220 ] relator families @xmath87 ( indicated in the figure by boxed numbers ) are used to eliminate @xmath96 .
the relator family @xmath88 expresses @xmath97 in terms of @xmath98 ; it also expresses @xmath99 in terms of @xmath100 .
the top front relator families are used to eliminate @xmath101 .
relator families @xmath102 are used to eliminate @xmath103 .
the relator family @xmath104 is regarded as redundant .
the bottom front relator families are of the form @xmath105 .
introduce a new generator family @xmath106 and defining relators @xmath107 .
rewrite the relators as @xmath108 , and use them to eliminate @xmath85 .
relator family @xmath88 can be used to eliminate @xmath109 for @xmath110 , writing them in terms of @xmath111 .
none of the remaining relators contain @xmath112 , and so these generate a free factor @xmath113 of rank @xmath24 .
the other generators and remaining relators are identical with those of the presentation obtained for @xmath46 .
hence @xmath64 is isomorphic to the free product @xmath114 .
we give examples to demonstrate the applicability of theorem [ alternate ] .
we show that the knot @xmath115 ( with braid index @xmath116 ) gives two @xmath116-braided fibered patterns and infinitely many mutually non - conjugate @xmath116-braided non - fibered patterns .
let @xmath117 denote the @xmath116-braid @xmath118 , where @xmath119 .
let @xmath120 be its closure along an axis @xmath121 .
each @xmath120 is ( the mirror image of ) the knot @xmath115 .
let @xmath122 denote a closed tubular neighborhood of @xmath121 that is disjoint from @xmath120 .
then @xmath123 is a solid torus , and @xmath124 is a pattern of @xmath115 .
[ 3componentexample ] with @xmath117 defined as above : 1 . the pattern @xmath124 is not fibered for any @xmath125 ; 2 .
patterns @xmath126 and @xmath127 are fibered . since @xmath128 is a homogeneous braid , j. stallings @xcite and l. rudolph @xcite claim that @xmath126 is a fibered pattern .
however , for completeness we include a proof for this case .
then encircled closures of @xmath128 and @xmath129 are links with distinct 1-variable alexander polynomials , and hence the braids are non - conjugate .
the braids @xmath117 , @xmath130 , will be seen to be mutually non - conjugate as a consequence of the fact that the genera of the @xmath3-component links consisting the closed braids and two copies of axes with opposite orientations are distinct .
consequently , the patterns @xmath124 are seen to be pairwise distinct .
_ for any @xmath119 , let @xmath131 denote the @xmath3-component link consisting of @xmath120 and @xmath132 , two copies of @xmath121 with opposite orientations .
we will construct a minimal genus seifert surface for @xmath131 , and show that it is not a fiber surface unless @xmath133 .
we use sutured manifold theory .
( for basic facts about sutured manifolds , see ( * ? ? ?
* 10 and appendix a ) and ( * ? ? ?
* section 1 ) . )
consider the seifert surface @xmath134 for @xmath135 as in figure [ a1 ] ( a ) , obtained by smoothing the ribbon intersections of the seifert surface for @xmath136 and the annulus spanned by @xmath132 . recall that if a seifert surface has minimal genus or is a fiber , then it remains so after plumbing or deplumbing of hopf bands
let @xmath137 be the surface as in figure [ a1 ] ( b ) , obtained from @xmath134 by deplumbing hopf bands .
-component links , height=1 ] consider the sutured manifold @xmath138 , where @xmath139 is an interval . apply a @xmath140-procut decomposition at each ribbon holeto obtain the sutured manifold as in figure [ a2 ] ( a ) .
the annulus part of @xmath134 gives rise to the four @xmath16-handles .
we can slide each of them as in figure [ a2 ] ( b ) and then remove them by @xmath140-product decompositions .
-product decomposition , height=2 ] we now have a sutured manifold which is a @xmath3-ball with a single suture . by @xcite , @xmath134 is a fiber surface ( of genus 5 ) and hence @xmath135 is fibered . by theorem
[ alternate ] , @xmath141 is a fibered pattern .
assume that @xmath142 .
let @xmath143 be a seifert surface for @xmath120 ( not of minimal genus ) as in figure [ a3 ] . by superimposing an annulus spanned by @xmath132 and smoothing the ribbon intersections as before , we obtain a seifert surface @xmath113 ( of genus @xmath144 if @xmath145 ) for @xmath131 .
recall that for any fibered link @xmath146 , fiber surfaces are unique seifert surfaces of minimal genus .
therefore , to show that @xmath131 is not a fibered link , it suffices to show that @xmath113 is not a fiber surface but that it has minimal genus . without loss of generality
, we can remove some of parallel half - twisted bands by deplumbing hopf bands so that the remaining ones are as in figure [ a4 ] ( a ) .
then we slide a band as in figure [ a4 ] ( b ) and remove the band marked @xmath147 by another hopf deplumbing .
obtain a sutured manifold by thickening the resulting seifert surface and apply @xmath140-product decompositions at each cite of a ribbon singularity .
the sutured manifold appears in figure [ a5 ] ( a ) .
as before , we can remove three @xmath16-handles marked @xmath147 by @xmath140-product decompositions . then applying another @xmath140-product decomposition along the shaded disk ( where the dots indicate the intersections with the sutures ) , we obtain a sutured manifold as in figure [ a5 ] ( b ) .
-product decompositions , height=2 ] a final @xmath140-product decomposition yields a sutured manifold @xmath148 . since it is a solid torus with four meridional sutures ( figure [ a5 ] ( c ) ) , its complement is not a product sutured manifold . by @xcite ,
@xmath113 is not a fiber surface .
however , we can apply a disk decomposition to @xmath148 so that the result is a @xmath3-ball with a single suture , and hence by @xcite , @xmath113 is a minimal genus seifert surface .
therefore , @xmath149 is not a fibered link . by theorem [ alternate ] , we conclude that @xmath124 is not fibered for any @xmath142 . in the remaining case @xmath150
, a sequence of @xmath140- product decompositions as above produces a solid torus with two meridional sutures .
consequently , @xmath151 is a fiber for @xmath152 .
we will denote by @xmath153 the collection of all fibered patterns @xmath2 such that @xmath0 is a trivial knot with winding number @xmath24 in @xmath1 .
the following theorem classifies the members of @xmath153 .
[ classify ] let @xmath2 be a pattern such that @xmath0 is the trivial knot with winding number in @xmath1 equal to @xmath24
. then @xmath2 is fibered if and only if the meridian @xmath41 of @xmath1 is the closure of an @xmath24-braid @xmath154 with axis @xmath0 .
the mapping @xmath155 $ ] is a bijection between @xmath153 and the set of conjugacy classes of the @xmath24-braid group @xmath156 .
_ we use notation established in section 2 .
assume that @xmath2 is a member of @xmath153 .
then there exists a fibration @xmath56 inducing the homomorphism @xmath47 on fundamental groups .
let @xmath157 be a fiber @xmath158 .
since @xmath159 restricts to a fibration of @xmath160 , it follows that @xmath161 consists of a longitude of @xmath0 together with longitudes @xmath162 of @xmath1 .
we can extend @xmath159 to a fibration @xmath163 such that @xmath164 is the capped off " fiber @xmath165 , since @xmath1 is unknotted .
assume now that @xmath0 is a trivial knot .
then @xmath164 must be a disk .
moreover , @xmath166 is isotopic in @xmath167 to a standardly embedded solid torus @xmath168 .
since @xmath41 intersects each @xmath169 in exactly @xmath24 points , @xmath41 is the closure of an @xmath24-braid @xmath154 with axis @xmath0 .
the particular embedding of @xmath41 in @xmath168 that we obtain in this way is independent up to isotopy of the choices that we make .
hence by theorem 1 of @xcite , the conjugacy class of @xmath154 in the @xmath24-braid group is well defined . in order to prove that the mapping @xmath155 $ ] is a bijection , it suffices to produce the inverse mapping .
the method that we use is essentially the reverse of the above , and it has been used elsewhere ( cf .
assume that @xmath154 is an @xmath24-braid that closes to a trivial knot @xmath41 .
denote the axis of @xmath41 by @xmath0 .
let @xmath170 and @xmath171 be disjoint tubular neighborhoods in @xmath167 of @xmath41 and @xmath0 , respectively , and consider the space @xmath172 .
regarded as the exterior of the closed @xmath24-braid @xmath41 in the solid torus @xmath173 , the space @xmath174 fibers over @xmath12 ; the fiber is a disk with @xmath24 holes and boundary consisting of a meridian of @xmath173 together with @xmath24 meridians of @xmath170 . on the other hand , @xmath174 can be viewed as the exterior of the trivial knot @xmath0 in the unknotted solid torus @xmath175 . from this vantage ,
the boundary of the fiber consists of a longitude of @xmath0 together with @xmath24 longitudes of @xmath1 , each also a meridian of @xmath176 .
it follows that the pattern @xmath2 is fibered .
notice that @xmath0 has winding number in @xmath1 equal to @xmath24 .
also , @xmath2 depends only on the conjugacy class of @xmath154 .
clearly , @xmath177 is the desired inverse mapping .
if @xmath41 is the the closure of a @xmath16-braid with unknotted axis @xmath0 , then @xmath40 is a hopf link .
we obtain the following corollary . in effect
, it asserts that there is no fibered satellite knot with unknotted pattern knot @xmath0 having winding number @xmath16 in the solid torus @xmath1 .
there exist nontrivial fibered patterns with winding number @xmath16 .
such an example @xmath179 can be constructed from the closure of the @xmath3-braid @xmath180 in @xmath181 by adding a short band with a full twist between the 1st and 2nd strings , away from the crossings . here
@xmath182 naturally spans a fiber surface @xmath53 consisting of a checkerboard - colorable surface together with a plumbed hopf band .
a fiber @xmath157 for the @xmath3-component link of theorem [ alternate ] is obtained by adding an annulus to @xmath53 . by theorem
[ alternate ] and arguments similar to earlier ones , we see that @xmath179 is a fibered pattern . by reasoning similar to that of corollary [ unknotted ]
we find that there exist only two fibered patterns @xmath2 such that @xmath0 is a trivial knot with winding number in @xmath1 equal to @xmath39 .
these correspond to the conjugacy classes of the @xmath39-braids @xmath183 and @xmath184 .
[ 2braids ] . for @xmath185 , there are only three fibered patterns , corresponding to the conjugacy classes of @xmath3-braids @xmath186 and @xmath187 @xcite .
( the braid @xmath188 does not appear in the list since it is conjugate to @xmath189 in @xmath190 . ) if @xmath24 is greater than 3 , the situation is more complicated .
we note that @xmath191 represent infinitely many pairwise non - conjugate @xmath116-braids @xcite .
axes @xmath41 of non - conjugate braids form distinct fibered patterns of the unknot by theorem [ classify ] . hence @xmath192 is infinite .
_ by theorem [ criterion1 ] it suffices to prove that @xmath0 can be embedded in @xmath1 with arbitrary winding number @xmath24 such that @xmath2 is not fibered .
we may assume that @xmath24 is positive ( see remarks preceding theorem [ criterion1 ] ) .
we will prove the corollary assuming further that @xmath0 is the trivial knot .
the general result will then follow by trying an arbitrary local knot " in @xmath0 ( that is , forming the knot product of @xmath0 and an arbitrary knot , and doing so within a @xmath3-ball in @xmath1 ) , since a product of knots is fibered if and only if each of the factors is .
let @xmath170 be a tubular neighborhood in @xmath167 of @xmath41 that is disjoint from @xmath0 .
then @xmath193 is an unknotted solid torus @xmath1 containing the trivial knot @xmath0 with winding number @xmath24 .
( when @xmath150 , the pattern @xmath2 is the configuration in fig . [ link ] ( b ) ( cf .
@xcite ) ) .
assume that @xmath2 is fibered . by theorem
[ classify ] , the link @xmath131 can be isotoped in such a way that @xmath41 appears as the closure of an @xmath24-braid with axis @xmath0 . consequently , @xmath131 is the closure of an @xmath194-braid . a theorem of j. franks and r.f
. williams @xcite ( also proved by h.r .
morton ) implies that the braid index of @xmath131 is at least @xmath195 , where @xmath196 and @xmath197 are respectively the largest and smallest exponents of @xmath198 in the @xmath39-variable jones polynomial @xmath199 of @xmath131 .
an easy calculation reveals that @xmath200 , where @xmath201 denotes the left - hand trefoil knot , and from this it is not difficult to verify that @xmath202 , a contradiction . hence @xmath2 is not fibered .
j. stallings , _ on fibering certain @xmath3-manifolds _ , in _ topology of @xmath3-manifolds _ , proc .
1961 topology institute univ .
georgia ( ed .
fort , jr . ) , 95100 , prentice - hall inc .
, englewood cliffs , nj . | necessary and sufficient conditions are given for a satellite knot to be fibered .
any knot @xmath0 embeds in an unknotted solid torus @xmath1 with arbitrary winding number in such a way that no satellite knot with pattern @xmath2 is fibered .
in particular , there exist nonfibered satellite knots with fibered pattern and companion knots and nonzero winding number . |
the recent discovery at the large hadron collider ( lhc ) of a spin zero resonance @xcite , most certainly the standard model ( sm ) higgs boson , have already given some clues of what kind of new physics we need to proceed beyond . besides , we know that the sm is not the last answer to the questions brought by the huge endeavor carried over the last three decades of researching in particle physics . some of the new challenges involve neutrino s mass and mixing , the dark matter problem , the matter - antimatter asymmetry , among others .
this motivates us to search for new broader models and theories for attacking these puzzles in the most simple and easiest way .
one such possibility is the enlargement of the gauge group symmetry , implying in new interactions and particle spectrum .
here we focus on one of these gauge extensions of the sm , which can be realized at some tev scale , the @xmath2 @xcite gauge group , which we call 341 for short .
the 341 model is built as a simple extension of the electroweak symmetry gauge group @xmath3 of sm .
it was first suggested in the work of voloshin @xcite , who considered only the leptonic sector .
further developments of the model including quarks and electroweak currents are presented in ref .
the most attractive feature of this model is the fact that the two chiralities of the lightest leptons are part of a fundamental representation of the gauge group @xmath4 , unifying each lepton family in a single multiplet .
also , the explanation of the family replication comes from anomaly cancellation which , along with qcd asymptotic freedom , requires that only three fermion families might be present in the spectrum , as in the @xmath1 ( 331 ) models @xcite . besides
, new particles with distinct signature are predicted to appear close to the electroweak scale , like doubly charged vectors and scalars , with some of them carrying two units of lepton number ( the so called bileptons ) , among other features also shared with the 331 models . in the current literature
there are 341 models with or without exotic electric charges @xcite . here
we deal with the 341 model with exotic electric charges , where we have reduced the scalar sector to only three scalar quadruplets , which are enough to generate the exact number of goldstone bosons needed to break the 341 symmetry to the @xmath5 . in turn , we have disposed of one scalar quartet and the scalar decuplet from the original version .
these would be responsible for giving mass for the leptons and some of the quarks , but in our scheme they are not mandatory , since we are going to generate those masses by effective operators , as in ref .
our approach poses a considerable reduction in the physical scalar spectrum , which is more suitable for phenomenological studies .
moreover , we compute the running coupling of the abelian gauge group , @xmath6 , and show that there exists a landau pole in this model , similar to the minimal 331 model @xcite . nevertheless , in this work the 341 model breaks into a 331 model with
right - handed ( @xmath7 ) @xcite , or a 331 model with left - handed heavy neutral fermions ( @xmath8 ) @xcite , and this means that such embedding of the lower symmetry into the 341 model implies a landau pole for these 331 models too .
the advantage of such embedding is that such models are known to possess cold dark matter candidates @xcite , which are automatically incorporated into the body of this 341 model .
besides , any mechanism that employs non - renormalizable effective operators suppressed by a scale much higher than tev is no longer an option in this scheme , which is the case of neutrino mass generation in some scenarios @xcite .
nonetheless , we will show that all fermions , including neutrinos , gain mass from such operators even when the suppression scale is as low as the tev scale gev was needed to implement his scheme , much higher than what we obtain for our model . ] .
this work is organized as follows : section [ sec1 ] , we introduce the 341 model with exotic electric charges . in section [ sec2 ]
, we discuss the perturbative limit of the model , where we compute the value of its landau pole , with the intent of measuring the energy scale at which the model loses its perturbative character . in section [ sec3 ] ,
the fermion mass spectrum is obtained , including the neutrinos .
finally , in section [ sec5 ] , we draw our conclusions .
the 341 model is a gauge extension of the electroweak gauge group of sm , that possesses 12 gauge bosons besides the known 8 gluons of strong interactions and the electroweak , @xmath9 , @xmath10 and @xmath11 of the sm .
since the strong gauge group @xmath12 is kept intact under the symmetry breakdown to @xmath5 , we are going to simply omit it throughout this work . in the following we present the field content of the 341 model and write up its scalar potential and the yukawa lagrangian , which are going to be relevant to our purposes in this work . in order to assign the fermionic content of the model
, it is useful to recall that the electric charge operator is written as a linear combination of the diagonal generators of the electroweak gauge group , @xmath13 .
this allows us to write the electric charge assignment for the fields in the fundamental representation of @xmath14 as , @xmath15 and similarly for the anti - fundamental representation , exchanging the sign of the diagonal generators of @xmath16 , @xmath17 , @xmath18 and @xmath19 . here ,
@xmath20 and @xmath21 are parameters to be fixed according to the field distribution in the quadruplets , and @xmath22 is the corresponding @xmath23 charge .
considering the sm content , which we should recover in the symmetry breakdown process , the first two components of the left handed lepton quartet should recover the known doublet structure , meaning that the first component is a neutrino and the second a negatively charged lepton .
moreover , as we are interested in accommodating the associated charge conjugate states for these leptons in the same multiplet , recovering the @xmath7 model in the first symmetry breaking of 341 , the third component of the quartet should be the charge conjugate of the neutrino and the fourth one should be the positive charged partner of the negative lepton , @xmath24 .
this association implies that @xmath25 , @xmath26 and @xmath27 , which reduces the operator in eq .
( [ cargas1 ] ) to , @xmath28 then , we can arrange all the leptons ( including right - handed neutrinos ) in quartets of @xmath29 symmetry as follows , @xmath30 where @xmath31 @xmath32 @xmath33 , @xmath34 and the symbol @xmath35 indicates the respective representation for the multiplet under the 341 symmetry .
for the first generation of quarks , we can assign the left - handed fields to the fundamental representation of the @xmath29 symmetry , while the right - handed ones transform trivially under this symmetry , @xmath36 where @xmath37 and @xmath38 are the up and down quarks , respectively , @xmath39 and @xmath40 are the new exotic quarks with electric charges , @xmath41 and @xmath42 , respectively .
the second and third families of left - handed quarks are arranged in the anti - quartet representation of the @xmath29 symmetry , and right - handed quarks in the singlet representation , @xmath43 where @xmath44 @xmath45 and @xmath46 are exotic quarks with electric charges @xmath47 and @xmath48 , respectively .
this arrangement in different representations for the quark families is a consistency requirement such that the model be free of anomalies .
next we explore the scalar content of the model .
originally @xcite , the 341 model was built with four scalar quartets in order to engender the symmetry breakdown to @xmath5 , give mass to all quarks and avoid mixing among ordinary and exotic quarks .
also , one scalar decuplet was introduced so as to generate charged lepton masses .
one of the purposes of this work is to show that we can diminish the scalar content and still have all desirable properties concerning the symmetry breakdown and the fermion masses ( including neutrinos ) . here
we present the minimal scalar content which will allow us to accomplish this goal , @xmath49 which transform as , @xmath50 , @xmath51 and @xmath52 .
the desired pattern of symmetry breakdown can be obtained if we assume that the following neutral components develop nontrivial vacuum expectation value ( vev ) , @xmath53 , @xmath54 and @xmath55 .
the vev @xmath56 is responsible for the first step in breaking the 341 symmetry to 331 , while @xmath57 breaks the 331 symmetry to 321 , the sm , and the final breaking to @xmath5 is provided by @xmath58 . in this way we can impose ,
@xmath59 gev . indeed , as we are going to show later
, @xmath60 and @xmath57 should be no more than few tev or we lose perturbativity , which makes this model interesting in the sense it can be promptly tested at lhc or discarded soon . the reason we choose the @xmath61 quadruplet to develop vev only in the third component is related to the fact that we do not want mixing among ordinary and exotic quarks on the yukawa lagrangian , which guarantees the usual ckm mixing in the quark sector .
of course this is only possible if we also add a new symmetry to the model , in our case this is going to be a @xmath62 discrete symmetry that we implement later on . with the scalar and fermionic content defined , we can write down the lagrangian of the model , invariant under the gauge symmetry and the additional discrete symmetry mentioned above . next , we add this symmetry and present the yukawa and scalar lagrangian of the model , also defining our gauge boson spectrum .
let us impose a discrete @xmath62 symmetry that will be suitable to avoid the mixing among ordinary and exotic quarks but also allow for an appropriate scenario for generating fermion masses through effective operators , as we will see later . the @xmath63 charges are @xmath64 @xmath65 @xmath66 , with @xmath67 , where @xmath68 is the identity element of the group .
we then assign @xmath62 charges to the 341 model fields given in table [ table1 ] . .@xmath62 symmetry transformation properties for the 341 fields .
those fields not present in this table transform trivially under this discrete group . [
cols="<,<,<,<,<,<,<",options="header " , ] in table [ tab : tab1 ] we have also shown the values of @xmath69 and @xmath70 when the exotic quarks , @xmath71 ( with @xmath72 ) , are decoupled , in which case we used the new value for the rge
abelian coefficient in eq .
( [ bi341 ] ) , @xmath73 .
we observe that the highest values for the landau pole scale are reached as the 331 symmetry breaking scale approaches that of 341 model , and can be as high as @xmath74 tev when the exotic quarks are decoupled , although perturbativity is lost about half that scale .
we do not go beyond @xmath75 tev because our results start becoming inconsistent in the sense that @xmath70 gets smaller than @xmath76 , in other words , we lose perturbativity at lower energies than 341 symmetry breaking scale .
this has a deep implication for such class of models , imposing that they must manifest somewhere close to the electroweak scale and may be promptly tested at lhc and/or ilc .
finally , the economical 331 model @xcite , along with an extra scalar triplet and three singlet scalars , can be embedded in the model here studied .
the same is true for the 331 model when the right - handed neutrino is exchanged by a new neutral fermion , @xmath77 @xcite .
this may have interesting implications for these models as well as for the 341 model developed here .
for the 331 models , it is automatic that they are not valid up to some arbitrary high energy scale , being limited to the landau pole of our 341 model , @xmath78 tev .
such a low cutoff scale is an obstacle for some neutrino mass generation mechanisms in these models @xcite , a problem we address here with an appropriate discrete symmetry and non renormalizable effective operators . in what concerns the advantages for the 341 model
, we can mention the fact that some of these models possess natural dark matter candidates @xcite , which can be explored in the context of our 341 model . besides , since the scalar resonance discovered at lhc
@xcite can be perfectly adjusted into the 331 models @xcite , its phenomenology can be recovered by our model as well , with some room for extra new gauge bosons and scalars to be tested at lhc .
besides , it is possible that the new neutral singlet fields ( under the 331 symmetry ) , may increase the number of dark matter candidates , a question to be investigated elsewhere . next we show that the fermion mass spectrum can be reproduced by means of non renormalizable effective operators , once the @xmath62 symmetry introduced in the last section can be implemented in the model .
now that we have established the highest energy scale that works as a cutoff for this 341 model , we can make use of the @xmath62 symmetry imposed through the assignments in table [ table1 ] in order to generate the observed fermion mass spectrum . in the original version of 341 model ,
this spectrum was obtained by means of four scalar quartets plus a decuplet .
the presence of a natural cutoff scale , @xmath74 tev , due to the existence of a landau pole at that scale , turns this scenario suitable for introducing non renormalizable effective operators to get realistic fermion masses and mixing without plaguing the model with too many scalar multiplets . some of the sm quarks get contributions to their masses from the following dimension-6
effective operators : @xmath79 where again , @xmath80 and @xmath72 label the family number ( or generation ) . considering these operators and the yukawa lagrangian , eq .
( [ yukawa ] ) , the up - type quarks mass matrix , in the basis @xmath81
@xmath82 @xmath83 , is written as , @xmath84 while the down - type quarks mass matrix , in the basis ( @xmath85 @xmath86 @xmath87 , is @xmath88 it is remarkable that , by choosing @xmath89 , all entries in these matrices are proportional to the sm symmetry breaking scale , @xmath90 gev , tuned by the dimensionless couplings . in other words , by appropriately tuning these couplings we can recover the sm results concerning quark masses and mixing without much effort .
for example , if we take the above matrices as diagonal ( just for the sake of illustration ) we have : @xmath91 where @xmath92 , @xmath93 , @xmath94 , @xmath95 , @xmath96 e @xmath97 are the masses for the up , charm , top , down , strange and bottom quarks , respectively . for @xmath98 tev and @xmath99 tev , and
considering the known quark masses , @xmath100 we obtain the following values for the respective yukawa couplings , @xmath101 , @xmath102 , @xmath103 , @xmath104 , @xmath105 , @xmath106 . in what concerns the exotic quarks ,
their masses are obtained purely from the yukawa lagrangian , eq .
( [ yukawa ] ) , providing the mass for @xmath107 and @xmath108 , @xmath109 it also leads to the exotic quarks mass matrices in the basis @xmath110 @xmath111 , @xmath112 and also in the basis ( @xmath113 @xmath114 , @xmath115 these matrices can be conveniently diagonalized , leaving no massless quarks in the spectrum . instead , just for illustration , we can assume them diagonal and conservatively using , as reference , the lower values obtained by the cms collaboration @xcite for a sequential fourth family ( @xmath116 gev and @xmath117 gev , which we use for the exotically charged quarks as well ) , imply that all yukawa couplings for the exotic quarks have to be approximately bigger than 0.45 .
as we saw earlier , charged leptons do not receive mass from yukawa lagrangian , but there is an effective dimension-5 operator that does this job , @xmath118 implying the following mass relation for charged leptons , @xmath119 this term provides an sm - like mass term for charged leptons once we assume @xmath120 , with @xmath121 playing the role of a yukawa coupling . considering the set of values for the mass scale parameter we have been using and plugging in the known masses for the leptons , we obtain the following values for the @xmath121 couplings : @xmath122 , @xmath123 , @xmath124 , which are comparable to those of sm . as for the neutrinos ,
the lowest order effective operator that engender their masses is dimension-9 , @xmath125 + + h.c.\ , , \label{numassop}\end{aligned}\ ] ] where we are supposing a diagonal basis for simplicity , yielding , @xmath126 with the scales as before , we see that neutrino masses are proportional to @xmath127 gev , which demands tiny values for @xmath128 for a neutrino of @xmath129 ev , not a great improvement if we compare with sm plus right handed neutrinos forming dirac mass terms , but it accounts for sub - ev neutrino masses as well .
this procedure has shown to be effective in producing the desired fermion mass spectrum in the 341 model .
nevertheless , there is an unwanted side effect that has to be circumvented , it concerns the presence of effective operators that would trigger fast proton decay .
the lowest order effective dimension-8 operator that leads to proton decay is , @xmath130 where an anti - symmetric color contraction is implicit .
this operator contains the following term , @xmath131 which yields the decay channel , @xmath132 .
the simplest way to avoid proton decay is to impose a discrete @xmath133 symmetry over the quark fields , such that , @xmath134 that guarantees the proton stability at any order without compromising any of the results we have obtained .
in this work , we have built a 341 gauge model with a lesser scalar content than usual .
the original version contains four scalar quartets and a decuplet , while we reduced this sector to only three scalar quartets . that was possible because we have computed the beta function for its abelian gauge coupling , showing that a landau pole exists at a scale not greater than @xmath135 tev .
this suggests that some underlying structure could emerge before that scale is reached , allowing us to use non renormalizable effective operators to provide the known mass spectrum for the fermions , which otherwise would need the presence of those extra scalars .
it is amazing that such a low scale constrains this class of model at the current energies reachable at colliders such as lhc and/or ilc .
besides , a reduced spectrum makes it most attractive for phenomenological studies , competing with supersymmetry , which has not shown any sign of existence till now .
it has to be remarked that the existence of a cutoff scale as low as few tev , may circumvent the long standing hierarchy problem .
actually , it would appear meaningless to address this problem once the theory enters into a strongly coupled regime at tev scale , where we no longer expect that the sm could be sensitive to . also , an interesting outcome of our scenario is that a 331 models with neutral fermions , when embedded in this 341 model , have to be faced as possessing this same landau pole , a result that is not obvious in this class of 331 models , where the landau pole is believed to appear much beyond the planck scale @xcite .
this may have some impact on the phenomenology of such models like the impossibility of implementing neutrino mass under the assumption of existing an arbitrarily high cutoff energy scale for these models . on the other hand ,
knowing the advantages of these models concerning their dm candidates and higgs phenomenology , we can guarantee the same outcome for our model , besides some new extra particles that can be investigated under the light of current experiments at lhc .
this work was supported by conselho nacional de desenvolvimento cientfico e tecnolgico - cnpq ( a.g.d .
grant 03094/2013 - 3 , c.a.s.p . grant 306923/2013 - 0 , p.s.r.s .
grant 305390/2012 - 0 ) , a.g.d is also supported by the grant 2013/22079 - 8 , so paulo research foundation ( fapesp ) , and coordenao de aperfeioamento de pessoal de nvel superior - capes ( p.r.d.p . phd program scholarship ) .
here we derive the mass spectrum for the scalar fields . in order to do
that we consider the following shift in the neutral scalars by their vevs , @xmath136 we then obtain the conditions for the minimum of the scalar potential in eq.([pot ] ) , @xmath137 with these constraints we can build the cp - even neutral scalars mass matrix in the basis ( @xmath138 ) , @xmath139 according to the results in section [ sec2 ] , it is natural to assume @xmath140 , which we will use to simplify the computation of eigenvalues and eigenvectors for the cp - even mass matrix above .
this diagonalization is performed by employing perturbation theory to the second order , leading to the following mass eigenvalues , @xmath141 where @xmath142 @xmath143 ^{2}}{4c_{1 } % \left[\lambda _ { 5}^{2}-\left ( \lambda _ { 1}-\lambda _ { 3}\right ) \left ( \lambda _ { 3}-\lambda _ { 1}+\sqrt{\left ( \lambda _ { 1}-\lambda _ { 3}\right ) ^{2}+\lambda _ { 5}^{2}}\right ) \right]},\ ] ] @xmath144 @xmath145 ^{2}}{4c_{3 } % \left [ \lambda _ { 5}^{2}+\left ( \lambda _ { 1}-\lambda _ { 3}\right ) \left ( \lambda _ { 1}-\lambda _ { 3}+\sqrt{\left ( \lambda _ { 1}-\lambda _ { 3}\right ) ^{2}+\lambda _ { 5}^{2}}\right ) \right ] } , \ ] ] while the respective eingenstates are obtained by considering first order perturbation theory only , @xmath146 observe that the state @xmath147 is a goldstone boson and @xmath148 is identified with the higgs boson , once it is the only neutral scalar to get mass at the electroweak scale , besides coming from a doublet under @xmath149 symmetry similar to the sm one .
the remaining scalars , @xmath150 and @xmath151 , are heavier than the higgs , with masses proportional to the 341 symmetry breaking scale .
the neutral cp - odd scalars , @xmath152 @xmath153 @xmath154 and @xmath155 , are all massless .
together with @xmath156 , they complete the set of five neutral goldstone modes eaten by the five neutral massive gauge bosons . as for the simply charged scalars
, their mass matrices can be divided into two .
the mass matrix in the basis ( @xmath157 , @xmath158 ) , is given by , @xmath159 which , upon diagonalization , leads to the mass eigenvalues , @xmath160 and the respective eigenvectors , @xmath161 the second mass matrix , in the basis ( @xmath162 , @xmath163 ) , is @xmath164 yielding the mass eigenvalues , @xmath165 and respective eigenvectors , @xmath166 in the above results @xmath167 and @xmath168 , together with @xmath169 and @xmath170 ( which are already massless eigenstates ) , are all massless and represent the goldstone bosons eaten by the eight simply charged gauge bosons , while the scalar fields , @xmath171 and @xmath172 are massive and remain in the physical spectrum .
concerning the doubly charged scalars , we have the mass matrix in the basis @xmath173 , @xmath174 that leads to a null eigenvalue , @xmath175 , and @xmath176 whose eigenvectors are , @xmath177 the scalar @xmath178 is obviously a goldstone boson eaten by the doubly charged gauge boson , while @xmath179 remains in the spectrum , being a characteristic signature in this class of models .
below we present the numerical values of couplings chosen to give a higgs mass of 126 gev as well as the masses of all scalars when @xmath180 tev . @xmath181 which yield the cp - even scalar masses , @xmath182 while the simply charged scalars acquired the following masses .
@xmath183 finally , the doubly charged scalar gets a mass , @xmath184 we have to keep in mind that the above particles have non - standard couplings to fermions and it is not straightforward to compare them with the existing bounds from colliders , although they are heavy enough to be safe and tested at the next lhc run
. a more careful and thorough analysis should be made elsewhere .
the effect of 341 to 331 symmetry breaking driven by the vev @xmath60 , causes fermionic quartets to decompose into triplets plus singlets .
the 341 model presented here contains as a subgroup the 331 model with neutral fermions in the third component of the lepton triplets .
there are two possibilities for this extra neutral fermion ( one for each family ) , it can be the partner right - handed neutrino of the left - handed neutrino in the triplet , or it can be a new neutral fermion not related to the ordinary neutrino . in order to see that ,
let us decompose the content of the 341 model into multiplets of 331 in the following way : * leptons @xmath185 where , from now on , the transformation properties under parentheses refer to the 331 symmetry . * first quark generation @xmath186 * second and third quark generations @xmath187 * scalars @xmath188 next , we present the breaking of 331 model promoted by @xmath57 that leads to the 321 sm content plus extra singlets . for the leptons we have , @xmath189 where @xmath190 and @xmath191 for the first quark generation we have , @xmath192 where @xmath193 and @xmath194 . for the second and third quark generations
, we have @xmath195 where @xmath196 and @xmath197 .
the scalar triplets decompose into one doublet and one singlet each , @xmath198 where @xmath199 @xmath200 @xmath201 @xmath202 @xmath203 and @xmath204 .
the scalar doublet @xmath205 plays the role of the higgs doublet in the sm .
so , we recover all the effective sm doublets and singlets with their respective quantum numbers , plus right - handed neutrinos , extra quarks and scalars .
the gauge bosons obtain their masses from the lagrangian , @xmath206 which , after spontaneous symmetry breakdown , leads to the charged ( and non - hermitian ) gauge boson masses , @xmath207 it also leads to the mass matrix for the neutral gauge bosons , in the basis ( @xmath208 @xmath209 @xmath210 @xmath211 ) , @xmath212 where @xmath213 .
this mass matrix has determinant equal to zero , which guarantees the existence of a massless gauge boson , that we can associate with the photon . diagonalizing the matrix using the simplifying ( and reasonable ) assumption @xmath214
, we obtain the neutral gauge boson masses , as given below : @xmath215 } { 8h_{w}\left ( 1 - 4s_{w}^{2}\right ) } \approx ( 2.2~{\mbox tev})^2\ , , \label{neutralgbmass}\end{aligned}\ ] ] with the respective eigenstates , @xmath216 , \\
z^{\mu } & = & c_{w}w_{3}^{\mu } -s_{w}\left [ \frac{t_{w}}{\sqrt{3}}\left ( -w_{8}^{\mu } -2\sqrt{2}w_{15}^{\mu } \right ) + \sqrt{1 - 3t_{w}^{2}}w_{x}^{\mu } % \right ] , \\ z^{\prime
\mu } & = & \frac{\sqrt{3}}{3}\frac{\sqrt{1 - 3t_{w}^{2}}}{\sqrt{% 1 - 4s_{w}^{2}}}\left ( \sqrt{h_{w}}w_{8}^{\mu } -2\sqrt{2}\frac{s_{w}^{2}}{% \sqrt{h_{w}}}w_{15}^{\mu } \right ) + \frac{s_{w}\sqrt{1 - 3t_{w}^{2}}}{\sqrt{% h_{w}}}w_{x}^{\mu } , \nonumber \\ \\ z^{\prime \prime \mu } & = & \frac{\sqrt{3}\sqrt{1 - 4s_{w}^{2}}}{\sqrt{h_{w}}}% w_{15}^{\mu } + \frac{2\sqrt{2}s_{w}}{\sqrt{h_{w}}}w_{x}^{\mu } , \end{aligned}\ ] ] in the above equations we have used the sine of the electroweak mixing angle written as @xmath217 , and defined @xmath218 , @xmath219 and @xmath220 . from these results
we see that @xmath221 , recovering the @xmath222 of the @xmath7 @xcite model , and the @xmath223 boson reproduces the well established massive neutral gauge boson of sm .
we then have an extra neutral gauge boson , @xmath224 , which is heavier than @xmath225 , whose phenomenology may easily be probed at lhc and/or the next collider generation .
g. aad et al .
( atlas collaboration ) , phys .
b 716 ( 2012 ) 1 ; s. chatrchyan et al .
( cms collaboration ) , phys .
b 716 ( 2012 ) 30 ; moriond proceeding contributions , http://moriond.in2p3.fr/qcd/2013/morqcd13prog.html .
j. g. ferreira jr , p. r. d. pinheiro , c. a. de s. pires , p. s. rodrigues da silva , phys .
d 84 ( 2010 ) 095019 ; see also : d. t. huong , l. t. hue , m. c. rodriguez , h. n. long , nucl . phys .
b 870 ( 2013 ) 293 ; j. g. ferreira jr , c. a. de s. pires , p. s. rodrigues da silva , a. sampieri , phys . rev .
d 88 ( 2013 ) 105013 ; vo quoc phong , hoang ngoc long , vo thanh van , phys . rev .
d 88 ( 2013 ) 096009 .
j. k. mizukoshi , c. a. de s. pires , f. s. queiroz , p. s. rodrigues da silva , phys . rev .
d 83 ( 2011 ) 065024 ; j. d. ruiz - alvarez , c. a. de s. pires , farinaldo s. queiroz , d. restrepo , p. s. rodrigues da silva , phys . rev .
d 86 ( 2012 ) 075011 .
a. alves , e. ramirez barreto , a. g. dias , c. a. de s.pires , f. s. queiroz , p. s. rodrigues da silva , phys . rev .
d 84 ( 2011 ) 115004 ; a. alves , e. ramirez barreto , a. g. dias , c. a. de s. pires , f. s. queiroz , p. s. rodrigues da silva , eur .
j. c 73 ( 2013 ) 2288 ; w. caetano , c. a. de s. pires , p. s. rodrigues da silva , d. cogollo , farinaldo s. queiroz , eur .
j. c 73 ( 2013 ) 2607 ; chong - xing yue , qiu - yang shi , tian hua , nucl .
b 876 ( 2013 ) 747 . | we build a gauge model based on the @xmath0 symmetry where the scalar spectrum needed to generate gauge boson and fermion masses has a smaller scalar content than usually assumed in literature .
we compute the running of its abelian gauge coupling and show that a landau pole shows up at the tev scale , a fact that we use to consistently implement those fermion masses that are not generated by yukawa interactions , including neutrino masses .
this is appropriately achieved by non renormalizable effective operators , suppressed by the landau pole scale .
also , @xmath1 models embedded in this gauge structure are bound to be strongly coupled at this same energy scale , contrary to what is generally believed , and neutrino mass generation is rather explained through the same effective operators used in the larger gauge group . besides , their nice features , as the existence of cold dark matter candidates and the ability to reproduce the observed standard model higgs - like phenomenology , are automatically inherited by our model .
finally , our results imply that this model is constrained to be observed or discarded soon , since it must be realized at the currently probed energy scale in lhc .
electroweak gauge model , effective theory , fermion mass 12.60.cn , 12.60.fr , 14.60.pq , 14.60.st |
the two gap superconductor mgb@xmath0 is considered to be a good candidate for superconducting devices due to its comparatively high critical temperature , its easy handling and cheap preparation .
presently , its use in josephson junctions is being investigated @xcite .
one of its peculiar properties is the presence of two distinct superconducting gaps existing on different bands at the fermi surface .
these bands are @xmath1 and @xmath2 bands arising from the boron @xmath3 and @xmath4/@xmath5 orbitals , respectively @xcite .
these two types of bands possess different parity with respect to reflection at the boron plane .
this different parity suppresses scattering and transitions between the two types of bands , which is thought to be the main reason for the exceptional stability of the two gaps against impurity scattering @xcite . in the vicinity of a josephson junction
this parity can be broken , if the two mgb@xmath0 electrodes are grown with a misorientation angle of the crystal @xmath6-axis directions on both sides of the junction .
the size of the misorientation angle would provide a means of tuning the strength of interband transitions between the two sides of the junction .
this is what we wish to explore in the present work . in particular , we want to study the differential conductance of such a junction and demonstrate how these interband processes show up in the tunneling conductance and in andreev reflections . due to the importance of josephson junctions in numerous applications
the josephson effect has been subject of intense theoretical study .
the current - phase relation as well as the oscillation of the electron transport in a voltage biased point contact known as the ac - josephson effect has been theoretically investigated for different situations @xcite including unconventional superconductivity @xcite , spin active magnetic barriers @xcite and the effect of pair breaking due to disorder @xcite .
it has been shown that within the quasiclassical theory interfaces can be described by effective boundary conditions @xcite that can be included by mainly two different approaches : the scattering - matrix approach and the transfer - matrix approach @xcite .
both are formally equivalent but the range of applicability is different and the problem under consideration can help to decide which one should be chosen . here , we are going to generalize the transfer - matrix approach to the case of a multiband josephson junction and use it to calculate the differential conductance of a mgb@xmath0 josephson junction with misoriented electrodes .
the effects of quasiparticle and josephson tunneling between two multiband electrodes with application to mgb@xmath0 have been theoretically discussed by brinkman et al . within an extended btk - model @xcite explaining convincingly the absence of a double gap structure in @xmath6-axis tunneling spectra .
novel josephson effects between multiband and singleband superconductors including the claim of a phase shift of @xmath1 between the different gaps on the multiband side have been theoretically discussed by agterberg et al .
@xcite showing the rich and interesting physical content of josephson effects in multiband systems .
the experimental observation of subharmonic gap structures in mgb@xmath0 single crystal junctions @xcite and mgb@xmath0/nb micro - constrictions @xcite due to multiple andreev reflections can be understood within a multiband transfer - matrix approach .
we will start in the next section with a description of the transfer - matrix model generalizing the approach to include the effect of multiple gaps . in the third section we will derive the current expression for both the equilibrium case without an applied voltage and the non - equilibrium case with an applied voltage . in the fourth section we will model the effective hopping amplitudes at the interface making use of some general considerations . in the fifth section we will show the results of our calculations while in the last section we will conclude .
the quasiclassical theory of superconductivity has been proven to be a powerful tool to consider spatially inhomogeneous problems in equilibrium and non - equilibrium , for example to calculate the local quasiparticle density of states around vortices or in the vicinity of boundaries .
however it is only valid within the quasiclassical limit @xmath7 and can therefore only describe situations with slowly varying fields and potentials ( on the scale of the coherence length ) .
the interface between two superconducting regions of different orientation or between a normal metal and a superconductor represents a strong pertubation on a lengthscale much smaller than the coherence length which is in principle out of the range of validity of the quasiclassical theory . to describe this strong pertubation within the quasiclassical limit one has to find effective boundary conditions that connect the solutions on both sides .
the first formulation of these boundary conditions for nonmagnetic interfaces has been found by zaitsev @xcite .
afterwards they have been generalized by millis et al .
@xcite for magnetic active interfaces and have been explicitly solved by shelankov @xcite for equilibrium problems and complemented by eschrig @xcite for non - equilibrium problems within the powerful riccati parametrization of the quasiclassical theory @xcite . to describe the complicated processes at the interface between two superconductors including multiple andreev scattering the transfer - matrix approach
has been proven to be as well suitable as the scattering - matrix approach but gives a more intuitive understanding .
to describe what is happening at the boundary between two superconductors with several fermi surfaces we will generalize in this work the transfer - matrix description of cuevas and fogelstrm @xcite and kopu et al .
@xcite for multiple fermi surfaces .
the basic idea of this approach is to consider two decoupled superconducting electrodes described by two electrode green s functions @xmath8 that are calculated in a left ( @xmath9 ) and a right ( @xmath10 ) half - space with an inpenetrable surface . in a second step
we allow virtual hopping processes between the two electrodes including the possibility of intraband and interband hopping .
this serial expansion is exactly summed up to infinite order using the technique known from the dyson formalism . as has been shown by kopu et al .
this pertubative transfer - matrix approach is equivalent to other methods and can be used to calculate interfaces with arbitrary transparency . in the following
we will repeat the steps described by cuevas and fogelstrm and we will discuss the consequences that arise from the multiband character of the superconducting pairing interaction . to describe the hopping processes we start with a phenomenological transfer hamiltonian @xmath11 where @xmath12 and @xmath13 denote band indices corresponding to a special fermi surface and introducing the effective hopping amplitudes @xmath14 describing hopping processes from band @xmath13 in the right electrode to band @xmath12 in the left electrode , and @xmath15 correspondingly . in this formulation
we have enlarged our hilbert space to include not only the band index but also the quantum number of the electrodes . the full green s function and the @xmath16-matrix
can be written as @xmath17 taking into account reflection ( @xmath18 and @xmath19 ) and transmission processes ( @xmath20 and @xmath21 ) .
the green s functions @xmath22 and @xmath16-matrices @xmath23 themselves are @xmath24-matrices in the enlarged band space where @xmath25 is the number of bands @xmath26 consisting of 2@xmath272-matrices in keldysh space , denoted by the check symbol @xmath28 the retarded , advanced and keldysh green s functions are 2@xmath272 nambu - gorkov matrices in particle - hole space while we will neglect the spin degrees of freedom since we are concerned with spin singlet systems and we assume only non - magnetic interactions . besides the full green s functions we also have to consider the green s functions of the unperturbed left and right electrodes , which take a diagonal form in the space of the two electrode quantum numbers , and we can write down the coupling parameters as off - diagonal elements in the same notation : @xmath29 since we want to neglect direct interband scattering in the decoupled left and right electrodes the self - energy and therefore also the unpertubed green s functions take diagonal form in band space @xmath30 we will see in the following that the full matrix structure of the @xmath31 in band space prevents us from an easy decoupling of the different matrix elements as it can be done in the space of the electrode quantum numbers . therefore we have to solve the full matrix problem in band space . with regard to the application to the two - band superconductor mgb@xmath0 ,
where interband scattering processes are suppressed @xcite , we will go on with these simplified diagonal electrode green s functions but of course it is straight forward to extend the considerations to a full matrix problem including interband scattering .
the transmission processes can now be summed up in form of a @xmath16-matrix equation with @xmath32 this equation has to be complemented by the dyson equation which reads @xmath33 using the first identity of eq .
( [ eq : t - matrix ] ) repeatedly we can write @xmath34 if we now employ the diagonal structure of the unperturbed green s functions und the off - diagonal structure of the hopping matrices in the space of the electrode quantum numbers we get the following set of decoupled equations for the components of @xmath35 . we note the diagonal ( reflection ) elements of the @xmath16-matrix as @xmath36 and we can write the off - diagonal ( transmission ) elements as @xmath37 now we can perform the quasiclassical integration since there are no functions with spatial arguments from both sides of the interface any more .
the quasiclassical @xmath38-integration reads @xmath39 where the matrix @xmath40 gives the square root of the density of states at the fermi level in the normal state .
this matrix is proportional to the identity matrix in keldysh space but has different elements on the diagonal in band space @xmath41 we can go on defining the quasiclassical transfer and hopping matrices as @xmath42 finally we can write down a set of equations defining the quasiclassical transfer matrices @xmath43 and @xmath44 where the fermi surface average @xmath45 has to be performed with respect to each different fermi surface sheet under consideration .
if we now take a look at the quasiclassical transport equation for each band @xmath12 where the strong perturbation at the interface enters as a source term on the right hand side @xmath46_\otimes^{\alpha } \nonumber \\ = 2 \pi v_{f , n}^{\alpha } \left [ \check{t}_{ii } ( \hat{p}_f , \hat{p}_f ) , \check{g}_{\infty , i } ( \hat{p}_f ) \right]_\otimes^{\alpha } \delta(\vec{r } -\vec{r}_c ) \label{eq : transporteq}\end{aligned}\ ] ] complemented by the normalization condition @xmath47 we can calculate the rapid change of the quasiclassical propagator at the interface by integrating over the source term in eq .
( [ eq : transporteq ] ) neglecting all the slowly varying fields on the left hand side .
we get for the difference of the quasiclassical green s function on the incoming and the outgoing trajectory evaluated directly at the boundary @xmath48^{\alpha } \nonumber \\ & & = - 2 \pi i \left [ \check{t}_{ii } ( \hat{p}_f , \hat{p}_f ) , \check{g}_{\infty , i } ( \hat{p}_f ) \right]_\otimes^{\alpha } \label{eq : goutmgin}\end{aligned}\ ] ] now it is possible to construct a solution for @xmath49 and @xmath50 that fulfills the boundary condition given by eq .
( [ eq : goutmgin ] ) and that also satisfies the normalization condition of the quasiclassical propagator given by eq .
( [ eq : normaliz ] ) : @xmath51 where we have omitted the angular dependencies .
taking @xmath52 as boundary values we can now construct the full quasiclassical propagator on incoming and outgoing trajectories in the vicinity of an interface with arbitrary transparency . in the following we will use it to calculate the quasiparticle density of states at the boundary , the josephson currents across the boundary and the currents at an applied voltage for a multiband @xmath53-wave superconductor .
since there are no andreev bound states at the interface without a sign change of the order parameter for reflected trajectories we do not expect dramatic changes of the pairing potential amplitude near the boundary . in this case
we can go on without a self - consistent calculation of the local pairing potential .
the current density can be calculated integrating the keldysh component of the quasiclassical propagator over the quasiparticle energy variable @xmath54 .
we can write the current component in the band @xmath12 as @xmath55 in the following we want to calculate only the current component across the surface on one side of the contact , e.g. the left side , so we can write @xmath56^{\alpha } \right\rangle_{\alpha , \phi_+ } \nonumber \\ & = & e n_f^{(\alpha ) } \int \frac{d \epsilon}{4 } \left\langle v_{f , n}^{\alpha } j_\epsilon^{k,\alpha } \right\rangle_{\alpha , + } \label{eq : jn}\end{aligned}\ ] ] where @xmath57 means an average on the fermi surface corresponding to the band @xmath12 with the restriction that only angles corresponding to outgoing trajectories with @xmath58 are taken into account . as we have seen before we can calculate the difference of the incoming and the outgoing propagator in the vicinity of the surface as @xmath59_\otimes^{k,\alpha } \right\ } \label{eq
: spectral_current}\ ] ] this expression can be further simplified in the equilibrium case where the keldysh propagator can be written as difference of the retarded and advanced green s function multipled by a temperature dependent distribution function . in the non - equilibrium case where we apply a current over the contact we have to calculate the time evolution of the transfer - matrix and get a time - dependent solution for the current . to calculate the currents we have to perform the fermi surface average as defined in the previous section . since within our model none of the green s functions
posses an angular dependence we only have to evaluate the following expression for the @xmath2 and for the @xmath1 band @xmath60 to proceed further we must take into account that the different bands posses different fermi surface geometries .
as we have shown in a previous work the three dimensional tubular network of the @xmath1 bands can be approximatly described by a half torus with a fraction of the two toroidal radii of @xmath61 while the nearly two dimensional isotropic fermi surfaces of the @xmath2 bands can be described by distorted cylinders @xcite . taking the interface to be orientated perpendicular to the @xmath62-plane direction we can write the fermi surface averages as @xmath63 and @xmath64
if we determine the values of @xmath65 and @xmath66 from band structure calculations @xcite and insert the value of @xmath67 we finally get the weighting factors for the two different bands as @xmath68 and @xmath69 now we can calculate the josephson current as a function of the phase difference between the gaps on the left and on the right hand side of the junction from eq .
( [ eq : jn ] ) taking the incoming and outgoing green s functions from eq .
( [ eq : gm ] ) and eq .
( [ eq : gp ] ) and evaluating the quasiclassical @xmath70-matrix as defined in eq .
( [ eq : tll ] ) .
since the josephson current is an equilibrium property the @xmath71-products reduce to simple matrix multiplications . to compare the quasiclassical results of the temperature dependence of the critical currents with a simplified model
we can write down a two band analogon to the ambegaokar - baratoff formula . here
we have to consider the weighted sum of two single band ambegaokar - baratoff formulas using the temperature dependence of the pairing potentials calculated selfconsistently from the two band gap equation : @xmath72 here @xmath73 is an effective band dependent transmission coefficient . for low transparencies
it can be calculated from the hopping amplitudes as @xmath74 .
if we apply a constant voltage across an @xmath75-@xmath76-@xmath75 josephson junction we will find that the phase difference increases linear according to the josephson relation @xmath77 where the josephson frequency is proportional to the applied voltage @xmath78 .
this results in an oscillating current across the junction and is known as the ac - josephson effect . in this case we have to solve the non - equilibrium time - dependent @xmath70-matrix equation .
first of all we will simplify the keldysh component of @xmath79_\otimes$ ] and we will show that the problem can be reduced to calculate only the retarded and advanced parts of the transmission components of the quasiclassical @xmath70-matrix @xmath80 and @xmath81 . then we will determine the time dependence of the transfer - matrix components mentioned above by expanding them into a fourier series .
as shown in the appendix the commutator can be transformed in such a way that we can write the spectral current density as @xmath82 now we have to find the time evolution of the transfer - matrix . since there is no further complication from the multiband character of our problem we can directly follow the steps described by cuevas and fogelstrm @xcite . starting with the transfer - matrix equation @xmath83 and fourier transforming it to energy space
@xmath84 leads to an algebraic equation to determine @xmath85 .
further one can show that the @xmath70-matrix allows a fourier expansion of the form @xmath86 so the problem of calculating the current can be reduced to the problem of calculating the fourier components @xmath87 for the retarded and advanced part of the transfer - matrix , that are equally defined as @xmath88 in the following we will use the same notation for the advanced and the retarded part of the @xmath70-matrix since they are described by the same equation .
we will also use the notation @xmath89 and @xmath90 .
but we have to keep in mind , that all @xmath87 are not only 2@xmath272-matrices in nambu - gorkov space but are also @xmath24-matrices in band space .
finally we can write down the algebraic @xmath70-matrix equation as @xcite @xmath91 where we have defined the matrix components as @xmath92 and also @xmath93 and finally @xmath94 here and in the following the notation @xmath95 stands for the diagonal and the functions @xmath96 and @xmath97 stand for the off - diagonal parts of the equilibrium green s function , evaluated at @xmath98 as @xmath99 the matrices @xmath100 are only evaluated for @xmath101 and we can write them as @xmath102 now the algebraic set of equations defined by eq .
( [ eq : algebraic ] ) can be solved by standard recursive techniques that we will briefly demonstrate .
a significant simplification can be achieved if we use the following identity @xmath103 that follows directly from the definition of the @xmath104 .
now we can proceed further by defininig some ladder operators @xmath105 as @xmath106 it can be easily verified that @xmath107 is a solution of eq .
( [ eq : algebraic ] ) for @xmath108 and therefore all @xmath109 with even @xmath110 vanish identically .
so we only have to calculate the odd ladder operators . inserting the above relations in eq .
( [ eq : algebraic ] ) leads to recursion formulas for the @xmath105 for @xmath111 or @xmath112 respectively : @xmath113 since the @xmath109 are getting constant for high values of @xmath110 we can start with @xmath114 and @xmath115 counting down to @xmath116 or up to @xmath117 . after solving the equations for @xmath118 we are now able to determine all @xmath109 by recursive application of the ladder operators . to find
the @xmath118 we have to solve two coupled @xmath119 matrix equations in the combined band and nambu - gorkov space : @xmath120 of course this procedure has to be applied to the retarded as well as to the advanced components of the transfer matrices .
the missing transfer matrices can be found using the general relation @xmath121 as for example described by cuevas and fogelstrm @xcite .
with this procedure the spectral weight of the non - equilibrium current across an @xmath75-@xmath76-@xmath75 contact for a given voltage @xmath122 and a given quasiparticle energy @xmath54 can be calculated . from this
we get the time dependent non - equilibrium current in a voltage biased contact as @xmath123 with the fourier components @xmath124 for the special case of an @xmath25-@xmath76-@xmath75 contact we can choose without loss of generality the right hand side as the normal conducting side with @xmath125 and therefore also @xmath126 . in this case
the ladder operators vanish and we only have to calculate the transfer matrix components @xmath118 . this can be understood since the ladder operators create repeated andreev reflections between the superconducting electrodes and in the case of a superconductor - normal conductor interface we only have single andreev processes .
can be modelled by calculating the overlap integrals of the @xmath127-orbitals and the @xmath3-orbitals of the boron atoms , which are responsible for the formation of the superconducting @xmath2 and @xmath1 bands of this material .
the angle @xmath128 measures the relative orientation of the two crystals on the different sides of the interface .
[ fig : orbitals ] ] one of the crucial features of the two gap superconductor mgb@xmath0 is the orthogonality of the wavefunctions belonging to the @xmath1 and @xmath2 bands , respectively . here
, we want to consider the case that the two mgb@xmath0 electrodes possess a certain misorientation angle of the crystal @xmath6-axis directions on both sides of the junction .
the misorientation angle allows to change the interband transmission between the two sides . in the following
we will introduce a simple model to determine the effective hopping amplitudes at the interface between these two misaligned regions .
the @xmath2 bands are formed by the in - plane @xmath2 bonds of the @xmath127 orbitals of the boron atoms , while the @xmath1 band is formed by the weaker @xmath1 bonds of the boron @xmath3 orbitals connecting the different boron layers . to model an interface between two electrodes with different orientation we have to calculate the overlap integrals between the different orbitals rotated against each other ( see fig .
[ fig : orbitals ] ) .
since all orbitals show rotational symmetry with respect to the @xmath129-axis it is sufficient to study only the rotation around one of the other axes , e.g. the @xmath130-axis .
we can write down the effective one - particle wave function for the @xmath4 , the @xmath5 and the @xmath3 orbitals , neglecting the radial part of the wave function , as follows @xmath131 where @xmath132 are spherical harmonics . from this
we can construct the @xmath133 orbitals as @xmath134 which correspond to the two @xmath135 and @xmath136 sub - bands , as we will call them in the following .
if we introduce rotated spherical coordinates we can define the following set of transformations between the unrotated and the rotated frame @xmath137 where @xmath128 is the misorientation angle ( see fig . [
fig : orbitals ] ) .
now we can calculate the overlap integrals for the different orbitals as @xmath138 where we have used the abbreviation @xmath139 .
the missing integrals can be found by simple symmetry considerations . if we now calculate the total transmission probability for one particle for example from the @xmath135-band into one of the other bands we get @xmath140 as to be expected .
we find the same total transmission propability for the other bands as well .
if we multiply the overlap integrals with a global transmission factor @xmath70 describing the overall reflection and transmission properties of the point contact we can write the angular dependent effective hopping amplitudes as @xmath141 we can write them in matrix form as @xmath142 the effective hopping amplitudes enter the transfer - matrix equation as described in section [ sec : tmapproach ] . in fig .
[ fig : transmission ] we show the angular dependence of the effective hopping amplitudes .
note , that interband transmission vanishes for @xmath143 , when there is no misalignment . on the other hand , for @xmath144 interband transmission becomes maximal , however , intraband transmission from the @xmath1 band into the @xmath1 band @xmath145 vanishes .
this peculiar behavior is a direct consequence of the orthogonality of the @xmath2 and @xmath1 bands .
in this section we will discuss the dependence of the local density of states , the critical current and the differential conductance on the misorientation angle @xmath128 between the two electrodes of a superconducting mgb@xmath0 point contact . in the following
we assume that the gap value in the vicinity of the point contact is not changed so it can be taken as the bulk value .
all energies will be normalized to the larger gap value @xmath146 while we will take the ratio of the two gaps to be @xmath147 .
@xcite bands while the dashed line shows the dos of the quasiparticles in the @xmath1 band .
both curves are normalized to the normal conducting dos in the corresponding bands .
[ fig : densityofstates ] ] first we will calculate the local quasiparticle density of states ( ldos ) in the vicinity of the point contact . in equilibrium without an applied voltage we can calculate the ldos in band @xmath12 as @xmath148 using the expressions of eqs .
( [ eq : gm ] ) and ( [ eq : gp ] ) to calculate the quasiclassical green s function on the incoming and the outgoing trajectory . in fig .
[ fig : densityofstates ] the ldos is shown for two different contacts . in the first case
we assume that both electrodes are equally orientated ( @xmath149 ) and we have no interband tunneling .
then the different bands remain decoupled and we find a standard @xmath53-wave spectrum in the @xmath2 bands as well as in the @xmath1 band . in the second case
we assume a maximum misorientation ( @xmath150 ) and we find peaks at the position of the @xmath1 gap in the @xmath2 band spectrum and vice versa due to interband hopping processes . in both cases
the transparency parameter @xmath70 was chosen to be @xmath151 . ] in fig .
[ fig : criticalc ] we present calculations of the temperature dependence of the critical current for the same parameters . for two equally orientated electrodes
we find a very good agreement of the quasiclassical results for low transparencies with the analytical approximation given by eq .
( [ eq : ambegaokar ] ) .
this is clear since for @xmath143 we have a decoupled system and the critical current can be understood as a weighted sum of two single gap critical currents , that are well described by the ambegaokar - baratoff formula . for a maximum misorientation of @xmath152
we find a deviation from the classical behaviour and a nearly linear decrease of the critical current over a wide range of temperatures . in both cases
we have taken the normalized density of states at the fermi energy to be @xmath153 and @xmath154 corresponding to band structure calculations for mgb@xmath0 .
@xcite the deviation from the ambegaokar - baratoff result seen in fig .
[ fig : criticalc]b can be attributed to interband transmission processes .
-@xmath76-@xmath75 junction calculated in a multiband model with parameters adapted to mgb@xmath0 ( e.g. @xmath155 ) and low transparency .
the angle @xmath128 gives the relative orientation of the two crystals on both sides of the interface and determines the effective strength of interband and intraband tunneling .
[ fig : current_lowt ] ] -@xmath76-@xmath75 junction calculated in a multiband model with parameters adapted to mgb@xmath0 ( e.g. @xmath155 ) and low transparency .
the angle @xmath128 gives the relative orientation of the two crystals on both sides of the interface and determines the effective strength of interband and intraband tunneling .
[ fig : conductance_lowt ] ] in the next figures we consider a point contact with an applied voltage .
the current @xmath156 and the differential conductance @xmath157 can be normalized using the conductance of the contact in the normal state @xmath158 .
employing eq .
( [ eq : jn ] ) and eq .
( [ eq : neqcurrent ] ) the normal state conductance for quasiparticles of the band @xmath12 can be calculated as @xmath159^\alpha\ ] ] and the total conductance is a sum over all band components @xmath160 . for high transparencies it shows a strong dependence on the misorientation angle @xmath128 while it becomes independent of @xmath128 in the limit of low transparency . in this limit also
the calculation of the non - equilibrium properties can be considerably simplified . in the case of low transparency
we can neglect multiple andreev reflections and use a simple model that can be deduced from eq .
( [ eq : neqcurrent ] ) expanding it for small values of @xmath161 and keeping only terms in second order . in this case only the @xmath70-matrix components @xmath118 have to be calculated and they reduce to @xmath162
. then all the different terms in eq .
( [ eq : neqcurrent ] ) can be summed up to @xmath163 where we have used that in equilibrium @xmath164 f_f ( \epsilon)$ ] with a fermionic distribution function @xmath165 .
furthermore we have written the quasiparticle density of states as @xmath166 $ ] . in fig .
[ fig : current_lowt ] we have calculated the current voltage characteristics for three different misorientation angles @xmath128 within this model .
the dotted and dashed lines show the contributions of the different bands where we have added the formally equivalent contributions from the two @xmath2 bands . for two equally orientated crystals on both sides of the interface ( @xmath149 ) we have no interband hopping amplitudes and therefore we find two distinct increases in the current stemming from the @xmath2 band contribution at @xmath167 and from the @xmath1 band contribution at @xmath168 .
this can be seen as a simple addition of two decoupled single band contributions with different gap sizes .
for @xmath169 we find three distinct increases , the first one for @xmath170 , the second one for @xmath171 and the last one for @xmath172 since the intraband as well as the interband hopping amplitudes become finite .
for @xmath152 we find a maximal increase in the tunneling current for @xmath171 and a smaller contribution at @xmath172 as expected from the angle dependence of the hopping amplitudes .
intraband processes between the @xmath1 bands do not exist for this orientation since the @xmath173 rotated @xmath3 orbitals are orthogonal . in fig .
[ fig : conductance_lowt ] we show the corresponding differential conductance as a function of voltage obtained by numerical differentiation of the current voltage characteristics in fig .
[ fig : current_lowt ] . here
the increase of the tunneling current due to tunneling between the high density of states at the gap edges results in distinct peaks in the differential conductance .
-@xmath76-@xmath75 junction calculated in a multiband model with parameters adapted to mgb@xmath0 ( e.g. @xmath155 ) and maximal transparency @xmath174 .
the angle @xmath128 gives the relative orientation of the two crystals on both sides of the interface and determines the effective strength of interband and intraband tunneling .
[ fig : current_nis ] ] in comparison to fig .
[ fig : current_lowt ] and fig .
[ fig : conductance_lowt ] the situation for point contacts with high transparency is totally different . here
the occurrence of andreev reflections dominates the current voltage characteristics .
first we want to consider an @xmath25-@xmath76-@xmath75 junction with multiband electrodes on both sides of the interface .
this situation could for example be experimentally realized by a reduced critical temperature in one of the electrodes due to defects .
andreev reflection processes carry twice the charge as the corresponding one electron process .
thus we would expect in the case of maximum transparency an increase of the differential conductance for low voltage leading to a value that is two times larger than the high voltage asymptotics .
this case is realized in fig .
[ fig : current_nis ] for @xmath143 and the results are in good agreement with the btk - results presented by brinkman et al . @xcite .
if we assume a misorientation of the electrodes we find a decrease of the differential conductance for low energies that can be explained by a reduction of the current in the @xmath1 band as a result of an effectively smaller transparency of the barrier .
-@xmath76-@xmath75 junction calculated in a multiband model with parameters adapted to mgb@xmath0 ( e.g. @xmath155 ) and high transparency @xmath151 .
the angle @xmath128 gives the relative orientation of the two crystals on both sides of the interface and determines the effective strength of interband and intraband tunneling .
[ fig : conductance_hight ] ] in the last figure ( fig .
[ fig : conductance_hight ] ) we show the differential conductance for an mgb@xmath0 @xmath75-@xmath76-@xmath75 point contact calculated within the non - equilibrium quasiclassical theory . for @xmath143
we find distinct conductance peaks at @xmath175 and @xmath172 and also the well known subgap structures at @xmath176 with @xmath177 that appear due to multiple andreev reflections .
if we rotate the electrodes we find a more complicated situation . for @xmath178
the peak at @xmath172 shrinks visibly as expected due to a reduced intraband hopping probability while the peak at @xmath179 increases .
this can be easily understood since not only hopping processes between the @xmath1 bands contribute to this peak but also multiple andreev reflections corresponding to @xmath180 and interband processes with @xmath181 . for @xmath152
the peak at @xmath172 has nearly vanished while we find at @xmath175 a dramatic reduction of the differential conductance .
instead we find an increase of interband processes corresponding to @xmath182 ( arrows ) .
this peak at @xmath183 is a characteristic signature of interband transmission .
it is absent for @xmath143 , when the electrodes are aligned , and it becomes maximum for @xmath150 .
an experimental observation of this effect would provide direct evidence for the orthogonality of the wavefunctions belonging to the @xmath1 and @xmath2 bands .
due to its multiband character mgb@xmath0 josephson junction promise to show rich and interesting physical properties .
we presented calculations of current - voltage characteristics of both @xmath75-@xmath76-@xmath75 and @xmath25-@xmath76-@xmath75 contacts .
we developed a simple model to determine the effective intraband and interband hopping amplitudes at the interface from a microscopic point of view .
we used these model parameters within a multiband transfer - matrix description based on the work of cuevas and fogelstrm and kopu et al . to study the dependence of the tunneling currents on the relative orientation of the two electrodes . especially due to the effect of multiple intraband and interband andreev reflections we were able to identify dramatic changes in the differential conductance for different electrode orientation . for @xmath25-@xmath76-@xmath75 junctions with equally orientated electrodes
our result agrees with the btk - results of brinkman et al . for @xmath75-@xmath76-@xmath75 junctions we found subgap structures at different voltages corresponding to different combinations of the two gaps .
the features sensitively depend on the misorientation angle of the junction .
we hope that especially the angular dependence of the differential conductance can be experimentally verified for point contacts on mgb@xmath0 single crystals which would provide direct evidence for the orthogonality of the two bands .
we would like to thank n. schopohl , c. iniotakis and a. gumann for valuable discussions .
this work was supported through the projektfrderung fr nachwuchswissenschaftler of the university of tbingen .
in this appendix we will show how the commutator formula for the spectral current density @xmath184 can be transformed into an expression that only contains the retarded and advanced components of the off - diagonal transfer matrix elements @xmath185 and @xmath186 .
first we show that the keldysh component of the @xmath16-matrix can be replaced by the corresponding advanced and retarded components . in the following
we will show the relations for the gorkov green s functions but of course they are also valid for the components of the quasiclassical green s propagator . starting with eq .
( [ eq : t - matrix ] ) , writing it out in keldysh - space we get for the retarded and the keldysh components of the @xmath16-matrix the following relations @xmath187 and @xmath188 solving the first equation for @xmath189 and the second equation for @xmath190 we get the following expressions @xmath191 and also @xmath192 substituting the first equation into the second equation we end up with the important relation @xmath193 with this equation we are able to replace the keldysh component of the transfer matrix by the keldysh component of the decoupled green s function , that can be easily calculated within the quasiclassical theory . in a second step
we will replace the reflection component @xmath194 by the corresponding transmission components @xmath195 and @xmath196 . comparing the expressions for the transmission and reflection components we find @xmath197 and therefore @xmath198 to calculate the commutator appearing in the current formula we are interested in expressions containing @xmath199 and @xmath200 . to find them we start with @xmath201 and performing @xmath202 we are now able to make use of the very important relation @xmath203 and we get @xmath204 or @xmath205
now we are able to replace the left hand side of eq .
( [ eq : tll_tlr ] ) with the expression above and after removing @xmath206 we find @xmath207 with basically the same idea we can find a relation between @xmath194 and @xmath196 .
we start with @xmath208 multiplication of @xmath209 from the right and using again the relation @xmath203 leads to @xmath210 another multplication from the right , now with @xmath211 , let us compare the right hand side of the equation with the definition of @xmath195 and we find @xmath212 where the second equation has been found by interchanging @xmath9 and @xmath10 . with this connection between @xmath195 and @xmath196
we can finally find another useful relation by substituting the expression for @xmath195 from eq .
( [ eq : connect1 ] ) on the right hand side and removing the @xmath213 factor @xmath214 the spectral current density can be calculated by eq .
( [ eq : spectral_current ] ) . in the following
we will use the abbreviation : @xmath215_\circ^k\ ] ] if we evaluate the commutator and perform the matrix multiplications in keldysh space we get @xmath216 in the next step we can replace the keldysh components of the transfer - matrices by its corresponding retarded and advanced components employing the relation of appendix [ sub : app11 ] : @xmath217 now the commutator takes the form @xmath218 in the last step we will replace the reflection components of the transfer matrix by the corresponding transmission components using the relations of appendix [ sub : app12 ] .
we start by substituting the brackets using eqs .
( [ eq : connect1 ] ) and ( [ eq : connect2 ] ) @xmath219 and @xmath220 the remaining two reflection components can be eliminated employing eq .
( [ eq : tll_tlr ] ) and eq .
( [ eq : connect4 ] ) leading to @xmath221 and we can finally write down the commutator corresponding to the spectral current density as @xmath222 performing the quasiclassical limit this expression defines the spectral current as it is given by eq .
( [ eq : neqcurrent ] ) . | we study the influence of multiple bands on the properties of josephson junctions .
in particular we focus on the two gap superconductor magnesium diboride .
we present a formalism to describe tunneling at a point contact between two mgb@xmath0 electrodes generalizing the transfer - matrix approach to multiple bands .
a simple model is presented to determine the effective hopping amplitudes between the different energy bands as a function of the misorientation angle of the electrodes .
we calculate the critical current and the current - voltage characteristics for n - i - s and s - i - s contacts with different orientation for junctions with both high and low transparency .
we find that interband tunneling processes become increasingly important with increasing misorientation angle .
this is reflected in certain features in the differential tunneling conductance in both the tunneling limit as well as for multiple andreev reflections . |
ls i + 61@xmath0303 is among the few binaries observed to date to emit high energy @xmath4-rays .
it hosts a bo ve main - sequence star ( hutchings & crampton 1981 ; casares et al . 2005 ) , orbited by a compact object of yet unknown nature , with a period of [email protected] days ( gregory 2002 ) .
the system is located at a distance of about 2 kpc .
the orbital zero phase is taken at jd 2443366.775 ( gregory & taylor 1978 ) , and the phase values of [email protected] ( casares et al . 2005 ) , @xmath20.275 ( aragona et al . 2009 ) , and [email protected] ( grundstrom et al
. 2007 ) represent the periastron passage uncertainty .
ls i + 61@xmath0303 has been recently detected as a periodic @xmath4-ray source by the major atmospheric imaging cerenkov telescope ( magic , albert et al .
2006 ) and the very energetic radiation imaging telescope array system ( veritas , acciari et al .
a distinctive orbital modulation of the vhe @xmath4-ray emission was seen , which was found to be anti - correlated with that observed at gev energies by the _ fermi _
gamma - ray satellite ( abdo et al .
2009 ) . in the last few years
, there has been a burst of activity trying to understand the nature of this source , whether it is composed by a pulsar or a black hole system , and which are the mechanisms that lead to their multiwavelength behavior ( e.g. , dubus 2006 , gupta & bottcher 2006 , romero et al .
2007 , sierpowska - bartosik & torres 2009 ) .
the current observations on ls i + 61@xmath0303 are rather sporadic , especially at high energies , which prevent from monitoring the phase evolution on time scale of years .
the limited observations at soft x - rays conducted by _ xmm - newton _ ( neronov & chernyakova 2006 ; sidoli et al .
2006 ) , _ chandra _ ( paredes et al .
2007 , rea et al . 2010 ) , _ asca _ ( leahy et al . 1997 ) , _ rosat _ ( goldoni & mereghetti 1995 ; taylor et al . 1996 ) and _ einstein _ ( bignami et al . 1981 ) were too short to cover an entire orbit . an exception to this were the _
r_xte / pca observations .
_ r_xte observed ls i + 61@xmath0303 in 1996 march 1 30 , and found that the 2 10 kev flux peaks at the orbital phase @xmath2 0.45 0.6 .
the detection and non - detection by hexte suggest a similar trend in orbital modulation at 15 - 150 kev ( harrison et al .
smith et al .
( 2009 ) adopted about half year of _
r_xte / pca observations covering the time period from august 28 , 2007 to february 2 , 2008 ( mjd 54340 - 54498 ) and reported a possible two - peak orbital lightcurve at 210 kev band and several very short flares ( which due to the large fov of the pca instrument , could still come from other sources in the field ) . the orbital phase derived at soft x - rays ( 0.3 - 10 kev ) peaks at the phase 0.65 , close to the apastron ( esposito et al .
this is consistent with the contemporary observations at above hundred gev by magic ( albert et al .
2009,2009b ) . at hard x - rays
, although the source was monitored by _ integral _ ( chernyakova , neronov & walter 2006 ) , data analyzed till now was not enough to explore the appearance of the orbital period in the power spectrum nor to construct a detailed lightcurve . here we investigate the hard x - ray spectral and timing properties of this source using the ibis / isgri instrument on - board the _ integral _ and carry out a systematic analysis based on all available _ integral _ data since december 28 , 2002 up to april 30 , 2008 . the total exposure time analyzed amounts to 2.1 ms thus enhancing the previously published report ( chernyakova , neronov & walter 2006 ) by taking into account more than 3 years of additional data , more than doubling the previous sample .
_ integral _ ( winkler et al .
2003 ) is a 15 kev - 10 mev @xmath4-ray mission .
its main instruments are the imager on board the _ integral _ satellite ( ibis , 15 kev 10 mev ; ubertini et al .
2003 ) and the spectrometer onboard _ integral _
( spi , 20 kev - 8 mev ; vedrenne et al .
these instruments are supplemented by the joint european x - ray monitor ( jem - x , 3 - 35 kev ) ( lund et al .
2003 ) and the optical monitor camera ( omc , v , 500 600 nm ) ( mas - hesse et al .
2003 ) . at the lower energies ( 15 kev 1 mev )
, the cdte array isgri ( lebrun et al . 2003 ) of ibis has a better continuum sensitivity than spi .
the satellite was launched in october 2002 into an elliptical orbit with a period of 3 days . due to the coded - mask design of the detectors
, the satellite normally operates in dithering mode , which suppresses the systematic effects on spatial and temporal backgrounds .
the _ integral _ observations were carried out in the so - called individual science windows ( scws ) , with a typical time duration of about 2000 seconds each . in this work ,
only ibis / isgri public data are taken into account ; the source is too weak to be detected by jemx and spi .
the available _ integral _ observations , when ls i + 61@xmath0303 had offset angle less than 14 degrees , comprised about 875 scws , adding up to a total exposure time of @xmath22.1 ms ( covering rev .
25 667 , mjd : 52636 54586 ) .
this total exposure then enhances the previously published report ( chernyakova , neronov & walter 2006 ) by taking into account more than 3 years of additional data , i.e. , more than double the previous sample .
the data reduction was performed by using the standard online science analysis ( osa ) , version 9.0 .
the results were obtained by running the pipeline from the flowchart to the image level , and the spectrum was derived using the mosaic images , as are appropriate for spectral analysis of faint sources .
the spectra were fitted with xspec v12.3.1 and the errors on the model parameters were estimated at 90@xmath5 confidence level . in order to search for a periodic signal in the lightcurve data , we used the lomb - scargle periodogram method ( lomb 1976 ; scargle 1982 ) and followed the procedure described in farrell et al .
power spectra were generated for the lightcurve using the period subroutine ( press & rybicki 1989 ) .
we set oversampling and high - frequency factors , which determine the period range and resolution ( press & rybicki 1989 ) , to two and four , respectively , allowing to search for periodic variability in the 1.1 1000 d range .
the 99% white noise significance levels were estimated using monte carlo simulations ( see e.g. kong , charles & kuulkers 1998 ) .
the 99% red noise significance levels were estimated using the redfit subroutine , which can provide the red - noise spectrum via fitting a first - order auto - regressive process to the time series ( schulz & mudelsee 2002 , farrell et al .
2009 ) .
[ tab1 ] .the flux and significance of ls i + 61@xmath0303 obtained from the mosaic images of the combined isgri / ibis observations . [ cols="<,^,^,^ , < " , ] the most significant _ integral _ detection of ls i + 61@xmath0303 is derived from combining all the isgri data , and is found at a significance level of @xmath2 12.0@xmath1 in the 1860 kev band ( see figure 1 ) .
this significance is consistently higher than that reported by chernyakova , neronov & walter ( 2006 , of 8.1@xmath1 ) at the 2263 kev , based on 600 scws ( covering rev .
25 - 288 , from 2003 january to 2005 march , and producing a corrected exposure of 273 ks ; as opposed to ours of about 578 ks ) .
table 1 shows the details on the imaging results in 4 energy bands : 20 - 40 , 40 - 100 , 100 - 300 , and 18 - 60 kev .
the neighboring source qso 0241 + 622 is detected at 3.4 mcrab level in 18 - 60 kev , implying a @xmath6 detection .
this source is only about 1.4 degrees away and it is usually a problem for the analysis of ls i + 61@xmath0303 with non - imaging detectors . the energy spectrum of the source is produced combining all the isgri data in the 13 kev 1.0 mev band . since we note that the sky region around ls i + 61@xmath7 303 is rather crowded and ls i + 61@xmath0303 itself is relatively faint in hard x - rays , it is necessary to obtain the energy spectrum from the mosaic images ( see also chernyakova & neronov 2008 ) .
the energy spectrum is well fitted by a power law shape , with a photon index of [email protected] and normalization of 3.1@xmath8@xmath910@xmath10ph @xmath11 s@xmath12 kev@xmath12 .
the flux we derive in the 20 - 200 kev band is 2.8 @xmath910@xmath13ergs @xmath11 s@xmath12 .
the reduced @xmath14 for the fit is 0.4 under 7 dof .
the orbitally separated results of the spectral fitting can be seen in table 2 .
these results show that our spectral fittings are mostly consistent with those published previously by chernyakova , neronov & walter ( 2006 ) .
they also present a large improvement , especially for the phase interval 0.6 - 0.8 , where the error bar in the photon index is largely reduced from @xmath15 to @xmath16 .
303 sky region derived by combining all isgri data at the 18 60 kev band .
the strongest source is qso 0241 + 622 ( at the upper part of the image ) and the relatively faint one is ls i + 61@xmath0303 .
the significance level is given in the color scale .
the contours start at a detection significance level of 5 @xmath1 , with a step of 3 @xmath1 . ] . ]
the isgri lightcurve of the source is obtained in the 1860 kev range with a time bin following the scws ( see figure 2 ) .
a fit to this lightcurve with a constant produces a value of @xmath17 1.2 ( 1058/874 ) , indicating a deviation at a significance level of 4@xmath1 .
we then investigated whether there is a recognizable orbital variability in this data using timing analysis .
the orbital lightcurve is then made by using the radio period and a reference time at mjd=43366.275 ( gregory 2002 ) at the hard x - rays , so that the absolute orbital phases can be directly compared with those at other frequencies ( and we do this below ) .
we find that at the hard x - rays the maxima is around phase 0.5 - 0.6 whereas the periastron of the system is at around 0.25 - 0.3 ( casares et al .
2005 , grundstrom et al .
2007 , aragona et al .
the lomb - scargle power spectrum based on the isgri lightcurve is presented in figure 3 .
the maximum of this power spectrum is located at around 27 days with a power density of 12.8 and an error of 4 days based on monte carlo simulations .
this periodical signal matches the orbital period of 26.4960 days , as reported in radio ( gregory 2002 ) . to obtain the error in the period determination we have sampled the light curve that was used to produce the power spectrum .
the flux and flux error of each bin of this light curve are regarded as the mean and the deviation of a gaussian .
accordingly , for each bin , a monte carlo flux is obtained to form a new light curve . in the power density of each so - generated lightcurves , the peak around 26.5 days
the width of this peak distribution provides one sigma error of the measured orbital period 26.72 days from the real isgri ldata .
over - plotted in figure 3 are the white noise and the red noise both at 99% confidence level .
the comparison of our _ integral _ lightcurve with those obtained at other frequencies is shown in figure 4 , which is further discussed next .
303 orbital period of 26.496 days is indicated with an arrow . ]
we have carried out a systematic analysis based on all available _ integral _ data since december 28 , 2002 up to april 30 , 2008 .
the total amount of time analyzed was @xmath22.1 ms , thus enhancing the previously published report ( chernyakova , neronov & walter 2006 ) by taking into account more than 3 years of additional data , more than doubling the previous sample .
corrected exposures after cuts changed from 273 to 578 ks .
the difference in the amount of data analyzed allows for a much more detailed _ integral _ lightcurve to be constructed , as can be visually seen in panel 3 of figure 4 above , where we show the chernyakova , neronov & walter ( 2006 ) 20 - 60 kev lightcurve in the background , and our current results superposed .
our analysis of _
integral_/isgri data results in the first direct orbital period determination through a power spectrum analysis of the hard x - ray data of ls i + 61@xmath0303 , although not yet truly significant , and in a spectrum at hard x - rays which is similar to the previous report ( chernyakova , neronov & walter 2006 ) , but with detection significance improved to 12.0@xmath1 , meaning that the source did not show significant changes in hard x - rays along the last five years .
our analysis also confirms that in the case of ls i + 61@xmath0303 there is no high - energy cut - off found at energies below 100 kev , as it would be common in conventional accretion scenarios .
indeed , if the system is an accreting neutron star or black hole , one expects to find a cut - off power - law spectrum in the hard x - ray band with a cut - off energy normally at 10 60 kev for neutron stars ( e.g. , filippova et al . 2005 ) and at @xmath2100 kev for black holes ( mcclintock & remillard 2003 ) .
even when one can think of situations where the accretion disk is masked by a jet component , producing a featureless power law , if the accretion flow and the jet would give comparable contributions in the hard x - ray band , one would still expect at least a feature in the integral energy range , due to the disappearance of of contribution of the disk at high energies
. such is not observed , and thus , in case an accretion disk is present , its contribution should be sub - dominant with respect to the jet one .
the flux we derive in the 20 - 200 kev band is 2.8 @xmath910@xmath13ergs @xmath11 s@xmath12 .
if most of it comes from a jet , and since in the scenario being discussion the whole source activity would be powered by accretion , the luminosity ( @xmath18 erg s@xmath12 , of the order of the gamma - ray luminosity ) could only be explained by a large beaming factor . from @xmath250 mas resolution radio images of ls
i + 61@xmath0303 obtained with multi - element radio linked interferometer ( merlin ) , extended , apparently precessing , radio emitting structures at angular extensions of @xmath19 arcsec have been reported by massi et al .
( 2001 , 2004 ) .
however , recent very long baseline array ( vlba ) imaging obtained by dhawan et al .
( 2006 ) over a full orbit of ls i + 61@xmath0303 has shown the radio emission to come from angular scales smaller than about 7 mas ( which is a projected size of 14 au at an assumed distance of 2 kpc ) .
no large features or high - velocity flows were noted in any of the observing days , which implies at least a non - permanent jet nature .
changes within 3 hours were found to be insignificant , so the velocity of the outflow can not be much over 0.05@xmath20 .
the magic collaboration also conducted a radio campaign ( in concurrence with tev observations ) to test these results using again merlin in the uk , the european vlbi network ( evn ) , and the vlba in the usa ( albert et al .
the results obtained by radio imaging at different angular scales show that the size of the radio emitting region of ls i + 61@xmath0303 is constrained to be below @xmath26 mas ( @xmath212 projected au ) , and the presence of persistent jets above this scale is therefore excluded . as in the case of dhawan et al .
( 2006 ) , these observations have shown a radio - emitting region extending east - southeast from the brighter , unresolved emitting core . the outflow velocity implied by these observations is @xmath21 .
these values of @xmath22 would result too low to significantly increase the luminosity ( @xmath23 ) even if strong jet precession is assumed . finally , search for lines with the chandra observatory yielded to no evidence for absorption or emission features in the source averaged and time - resolved spectra ( rea et al .
2010 ) . putting the _ integral _ observations in this context , we find no obvious support for the accretion model in our data . as can be seen in figure 4 , the hard x - ray emission peaks in phase around 0.5 , and in order to compare with the orbital phase derived at @xmath4-rays , we have also generated the orbital light curves of _ r_xte / pca and _ r_xte / asm using the currently available observations .
the orbital peaks of the soft and hard x - rays are close in phase ( 0.4 - 0.6 ) but not co - aligned .
thus , with low statistics , the hard x - rays may appear following the overall orbital modulation trend of the soft x - rays ( e.g. see harrison et al .
we also show in figure 4 the lightcurves obtained by fermi ( abdo et al .
2009 , 8 months of observations since launch in june 2008 ) , and magic ( several orbits put together , along few years of observations , as described in albert et al .
2009 ) . direct inspection of these lightcurves seem to indicate that the hard x - ray component peaks in between the average soft x - ray and tev lightcurves , but significant variability ( e.g. , orbit to orbit variability is already seen in magic and veritas data , albert et al .
2009 , acciari et al .
2009 ) may preclude to obtain conclusions in non - simultaneous observations .
this work was subsidized by the national natural science foundation of china , the cas key project kjcx2-yw - t03 , and 973 program 2009cb824800 . j .-
thanks the natural science foundation of china for support via nsfc-10325313 , 10521001 and 10733010 .
dft and nr acknowledges support from grants aya2009 - 07391 and sgr2009 - 811 .
abdo a. et al . , 2009 ,
apj , 701 , 123 acciari v.a .
et al . , 2009 ,
apj , 700 , 1034 albert j. et al . , 2006 , science , 312 , 1771 albert j. et al . , 2008 ,
apj , 684 , 1351 albert j. et al . , 2009 , apj , 693 , 303 albert j. et al . , 2009b , apj letters , 706 , 27 aragona c. et al . , 2009 ,
apj , 698 , 514 scargle j. d. , 1982 , apj , 263 , 835 schulz m. , mudelsee m. , 2002 , comput . geosci . , 28 , 421 sidoli l. et al . , 2009 ,
a&a , 459 , 901 sierpowska - bartosik a. , torres d. f. , 2009 , apj , 693 , 1462 smith a. et al . , 2009 ,
apj , 693 , 1621 | ls i + 61@xmath0303 is one of the few high - mass x - ray binaries that have been recently observed at tev and gev energies .
here we investigate the hard x - ray spectral and timing properties of this source using the ibis / isgri instrument on - board the _ integral _ satellite .
we carry out a systematic analysis based on all available _ integral _ data since december 28 , 2002 up to april 30 , 2008 . the total exposure time analyzed amounts to 2.1 ms , hence more than doubling the previous reported sample .
the source is best detected in the 18 - 60 kev band , with a significance level of 12.0@xmath1 .
the hard x - ray data are best fit with a simple power law with a photon index of @xmath2 [email protected] .
we detect a periodical signal at 27 @xmath3 4 days , matching the orbital period of 26.496 days previously reported at other wavelengths . the hard x - rays orbital lightcurve is obtained and compared with those derived at other frequencies .
[ firstpage ] x - rays : individuals : ls i + 61@xmath0303 |
avoided crossing of energy levels is a universal phenomenon for the quantum non - integrable systems where the symmetry break leads to the splitting of degenerate energy levels forming a tiny energy gap . around the avoided crossing point of the two levels the landau - zener tunnelling ( lzt ) model
provides an effective description for the tunnelling dynamics under assumption that the energy bias of two levels undergoes a linear change with time@xcite .
it is a basic model in quantum mechanics and has versatile applications in quantum chemistry @xcite , collision theory @xcite , and more recently in the spin tunnelling of nanomagnets wensdorf1,wensdorf2 , bose - einstein condensates ( bec ) @xcite and quantum computing @xcite , to name only a few .
lzt model has been extended to many versions taking diverse physical conditions into account : lzt problem with a time - varied sweeping rate@xcite , lzt model with a fast noise from the outer environment @xcite , lzt model with periodic modulation @xcite , and so on . among them , lzt in a nonlinear two - level system is one of most interesting models and attracts much attention recentlyraghavan , bwu , zobay , jliu1 . in this model ,
the level energies depend on the occupation of the levels , may arise in a meanfield treatment of a many - body system where the particles predominantly occupy two energy levels . the nonlinear lzt model
not only demonstrate many novel behavior of great interest in theory but also has important applications in spin tunnelling of nanomagnets @xcite and a bose - einstein condensate in a double - well potential @xcite or in an optical lattice @xcite .
however , since most of the problems of interests involve more than two energy levels , with transitions between several levels happening simultaneously@xcite , for example , becs trapped in multiple wells@xcite , spin tunnelling of nanomagnets with large spin , etc .
it is naturally desirable to extend the above nonlinear tunnelling to the multi - level situation . in present paper
, we consider the simplest multi - level system three - level system , to investigate its complicated tunnelling dynamics in the presence of nonlinearity .
because quantum transitions may happen between several levels simultaneously , the lzt in the nonlinear three - level model show many striking properties distinguished from that of the two - level case . in the adiabatic limit
we will show that , for a very small nonlinear parameter that the energy levels still keep the same topological structure as its linear counterpart , the adiabaticity breaks down manifesting the presence of a nonzero tunnelling probability .
this is quite different from the two - level case , where the break down of the adiabaticity is certainly accompanied by a topological change on the energy levels .
more interestingly , the tunnelling is irregular with showing an unresolved sensitivity on the sweeping rate , a phenomenon attributed to the existence of chaotic state . in the sudden limit
, we derive an analytic expression for the tunnelling probability under stationary phase approximation and show that the nonlinearity can dramatically influence the tunnelling probability at the resonance between the nonlinear `` internal field '' and the external field .
we also discuss the asymmetry of the tunnelling probability induced by the nonlinearity . the physical mechanism behind these phenomena is revealed and possible application of our model to triple - well trapped bose - einstein condensate is discussed .
the paper is organized as follows . in sec.ii
we introduce our nonlinear three - level lzt model and calculate its adiabatic levels .
section iii discusses lzt among the levels .
section iv gives a possible application of the model to the triple - well trapped bec .
we consider following dimensionless schrdinger equation @xmath0 with the hamiltonian given by @xmath1where @xmath2 is the coupling constant between the neighboring levels ; @xmath3 is the nonlinear parameter ; the energy bias @xmath4 is supposed to be adjusted by a linearly external filed , i.e. , @xmath5 , @xmath6 is the sweeping rate ; @xmath7 , @xmath8 , @xmath9 is probability amplitude in each level and the total probability @xmath10 is conserved and set to be unit .
when the nonlinear parameter vanishes , our model reduces to the linear case and the adiabatic energy levels @xmath11 ( fig.1(a ) ) derived by diagonalizing the hamiltonian ( 2 ) . tunnelling probability @xmath12 @xmath13
is defined as the occupation probability on the m - th level at @xmath14 for the state initially on the n - th level at @xmath15 . for the linear case ,
the above system is solvable analytically and the tunnelling probabilities can be explicitly expressed as @xcite @xmath16 ^{2 } \label{euqation3}\ ] ] @xmath17 \label{euqation4}\ ] ] @xmath18 @xmath19 ^{2 } \label{euqation6}\ ] ] the others are @xmath20 due to the symmetry of the levels . with the presence of the nonlinear terms ,
we want to know how the tunnelling dynamics in the above system is affected . in our discussions ,
the coupling parameter is set to be unit , i.e. , @xmath21 .
therefore , weak nonlinear case and strong nonlinear case mean that @xmath22 and @xmath23 , respectively . as to the external fields
, we will consider three cases , namely , adiabatic limit , sudden limit , and moderate case , corresponding to @xmath24 , @xmath25 and @xmath26 , respectively .
similar to the linear case , we need to analyze the adiabatic levels of the nonlinear model first . with @xmath27
, we introduce the relative phase @xmath28 . in terms of @xmath29 and @xmath30
, the nonlinear three - level system is casted into a classical hamiltonian system , @xmath31 @xmath32 @xmath29 and @xmath30 are two pairs of canonically conjugate variables of the classical hamiltonian system .
the fixed points of the nonlinear classical hamiltonian correspond to the eigenstates of the nonlinear three - level system , and are given by the following equations : @xmath33 @xmath34 @xmath35 @xmath36 by solving the equations ( [ euqation7])-([equation 10 ] ) the eigenstates of the system are obtained .
accordingly , the eigenenergy is obtained by @xmath37 i.e. , the energy levels are gained as shown in fig.1 . for weak nonlinearity ,
the levels structure is similar to its linear counterpart ( fig.1(b ) ) . for strong nonlinearity ( fig.1(c ) ) , in the mid - level a double - loop topological structure emerges and in the upper - level a butterfly structure appears . because of these topological distortions on the energy levels , we expect that the tunnelling dynamics will dramatically change .
in this section we study lzt in the nonlinear three - level system both numerically and analytically .
first , we consider two limit cases : adiabatic limit , sudden limit , respectively
. then we will discuss the tunnelling probability in general case and investigate the symmetry of the tunnelling probability . in adiabatic limit
, the characters of the tunnelling probabilities should be entirely determined by the topology of the energy levels and the eigenstates properties ( corresponding to the stability of the fixed points in classical hamiltonian system ) , according to the adiabatic theorem@xcite .
so , we expect that , for the weak nonlinearity case , an initial state started from any levels ( upper , mid or lower ) will follow the levels and evolves adiabatically , as a result , no quantum transition between levels occurs ; for the strong nonlinearity , an initial state from the lower level is expected to evolve adiabatically keeping stay on the ground state , leading to zero adiabatic tunnelling probability , whereas for the state initially from the mid or upper level , due to the topological change of the level , it can not move smoothly from left - side to the right - side .
transition to other levels happens at the tip of the loop or butterfly .
consequently , the adiabatic tunnelling probability is expected to be nonzero .
however , the above picture is only partly corroborated by our directly solving the schrdinger equation using forth - fifth order runge - kutta adaptive - step algorithm , as shown in fig.2 .
on the one hand , fig.2 clearly shows that , for the strong nonlinearity case , as we expect , no tunnelling for the state from the lower level , but a serious adiabatic tunnelling is observed for the states from upper two levels .
in particular , we find that the tunnelling probability as a function of the sweeping rate shows an irregular oscillation .
we associate this irregularity to the chaotic state . to demonstrate it , we plot in fig.3 the poincare section of the trajectories for @xmath39 before and after the tip of the butterfly structure of the upper level in fig.1c .
it shows that , before the tip , the eigenstate corresponds to the fixed point surrounded by quasi - periodic orbit , therefore is stable .
as the state evolves to the right tip of the butterfly , it contact with chaotic sea , after that the state become chaotic .
the characteristics of the chaos is sensitive on the parameters , therefore the chaotic state is responsible for the irregular tunnelling probability exposed by fig.2h , i . on the other hand , fig.2 also shows that for the weak nonlinearity , even though the adiabatic levels keeps the same topological structure as the linear case , there is still nonzero tunnelling probability for the state started from the mid - level .
the tunnelling also shows some kind of irregularity .
this phenomenon counter to our naive conjecture from observing the topological structure of the adiabatic levels . to explain this unusual phenomenon , we need make detailed analysis on the property of the fixed points of the classical system hamiltonian(7 ) , corresponding to the eigenstates of the mid - level .
we plot quantity @xmath40 as the function of @xmath41 in fig.4 ( a , b ) , we see the adiabatic evolution of the eigenstate breaks down around @xmath42 due to the nonlinearity ( fig.4 ( b ) ) .
this adiabaticity breakage is caused by the change on the property of the fixed point corresponding to the eigenstate of the mid - level .
this is revealed by investigating the hamiltonian - jaccobi matrix obatined by linearizing the nonlinear equations ( [ euqation7])-([equation 10 ] ) at fixed points , @xmath43we solve the eigenvalues of @xmath44 for different @xmath41 and plot our results in fig.4c .
these eigenvalues can be real , complex or pure imaginary .
only pure imaginary eigenvalues corresponds to the stable fixed point , others indicate the unstable ones . in fig.4
( c),we can see the eigenvalues are complex number ( i.e. , their real parts are not zero ) around @xmath45 .
the corresponding fixed points are unstable . for other regions ,
the eigenvalues of @xmath44 are pure imaginary . therefore , even though no topological structure changes on the level structures , the instability of the fixed point corresponding to the mid - level leads to the breakdown of the adiabaticity manifesting the irregular nonzero tunnelling probability exposed by fig.2e in the adiabatic limit .
the above instability mechanism occurs for any smaller nonlinear perturbation .
let us make some analytic deduction as follows .
note that the fixed points of equations ( [ euqation7])-([equation 10 ] ) can be accurately calculated if @xmath46 @xmath47 @xmath48 @xmath49 for @xmath50 and @xmath51 @xmath52 for @xmath53 . by employing the perturbation theory using @xmath3 as small parameter , we can get the fixed points for small @xmath54 @xmath55 @xmath56 @xmath48 @xmath57 for nonlinear case . substituting them into equation ( [ euqation9 ] ) , we can obtain the eigenvalues of @xmath44 by solving the following quartic equation : @xmath58the useful quadratic discriminant is @xmath59 in linear case , @xmath60 , @xmath61 is always larger than zero , which means that the solutions for @xmath62 are pure imaginary , thus the fixed points are stable . for small @xmath3 ,
@xmath63 the real part of the solutions @xmath64 while the imaginary part @xmath65 as a result , the fixed point corresponding to mid - level becomes unstable around @xmath66 for any small nonlinearity , implying the break down of the adiabatic evolution of states on the mid - level .
the sudden limit corresponds to nonadiabatic lzt .
the tunnelling probability does not relate much to the structure of the levels . in this limit
a weak nonlinearity does not affect the tunnelling probability , however , a strong nonlinearity can dramatically influence the tunnelling dynamics . in this limit
, we can derive the analytical expression of the tunnelling probabilities using the stationary phase approximation ( spa ) . as a demonstration ,
we concentrate on the mid - level , i.e. to calculate @xmath68 which equals to @xmath69 . because of the large sweeping rate @xmath70
, a quantum state would stay on the initial level most of the time .
thus the amplitudes @xmath7 and@xmath71 in the schrdinger equation ( 1 ) remain small and @xmath72 all the time .
a perturbation treatment of the problem becomes adequate .
we begin with the variable transformation , @xmath73,\]]@xmath74,\]]@xmath75.\]]as a result , the diagonal terms in hamiltonian are transformed away , and the evolution equations of @xmath76 , @xmath77 , @xmath78become : @xmath79\]]@xmath80 \\ & & -\frac{v}{2i}a_{3}^{^{\prime } } \exp [ i\int_{0}^{t}\left ( \frac{\gamma } { 2}+% \frac{c}{4}(\left| a_{2}\right| ^{2}-\left| a_{3}\right| ^{2})\right ) dt]\end{aligned}\]]@xmath81\ ] ] we need to calculate the above integrals self - consistently . due to the large @xmath70
, the nonlinear term in the exponent generally gives a rapid phase oscillation , which makes the integral small .
the dominant contribution comes from the stationary point @xmath82 of the phase around which we have @xmath83.\]]@xmath84with @xmath85 _ { t_{0}}.\]]since @xmath86 , then we have @xmath87 \right| ^{2}\]]this expression can be differentiated and evaluated its result at time @xmath88 . a standard fresnel integral with the result @xmath89 _ { t_{0}}=\left ( \frac{v}{2}\right ) ^{2}\sqrt{% \frac{\pi } { \alpha _ { 1}}}$ ] is obtained . combining this with the relation ( 15 ) ,
we come to a closed equation for @xmath90 , @xmath91for given @xmath70 , @xmath3 , @xmath2 , @xmath90 in the above equation can be obtained .
the tunnelling probability @xmath92 the alliance of eq.(20)and ( 21 ) gives the analytic expression on the tunnelling probability @xmath93 in the sudden limit .
compared with our numerical simulation it shows good agreement at @xmath94 , @xmath95 a clear deviation is observable ( fig.5a ) .
it is due to the resonance between the `` internal field '' and the external field leads to the invalidity of our assumption @xmath96 , as we show latter .
similarly , to calculate @xmath97 , we consider following equation , @xmath98\]]@xmath99@xmath100 differently , in this case we may have three stationary phase points that are solutions of equation ( 24 ) when @xmath101 , but only one solution otherwise , as demonstrated in fig.6 .
we denote them as @xmath102 , @xmath103 , @xmath104 from smallest to largest . for small c ,
@xmath102 is around @xmath105 , and the other two solutions locates at the two sides of the origin . in this case , we simply take @xmath106 . then @xmath107.\ ] ] the tunnelling probability @xmath108 the alliance of the eq.(24,26 ) will give the approximate solution of the @xmath97 .
compared with our numerical simulation it shows a good agreement at @xmath109 , whereas for @xmath110 a clear deviation is observed ( fig.5b ) . what happens around @xmath111 that leads to the break down of our stationary phase approximation ?
the reason is the resonance between the `` internal field '' and the external field .
let us recall the exponent in the integrand of eq.(22 ) , we find the effective sweeping rate should be the difference between the change rate of the `` internal field '' ( i.e. , @xmath112 ) and the sweeping rate of the external field . at @xmath111 , we find the two frequencies become almost identical , leading to the invalidity of spa assumption of rapid phase oscillation .
this resonance accompanied by the bifurcation of the stationary phase points . crossing @xmath111 we observe the number stationary phase points changes from three to one , as shown in fig.6 .
the resonance breaks the spa leading to serious transition from level 2 to level 1 , consequently , at @xmath95 , our assumption @xmath96 become invalid , and our approximation on the @xmath93 from spa is no longer good as shown in fig.5a .
the nonlinear tunnelling probability as the function of the two scaled quantities @xmath113 and @xmath114 , show many unusual properties . taking mid - level tunnelling @xmath68 for example
, we make a large numerical exploration for a wide range of parameters , to demonstrate the general property of the nonlinear tunnelling probability in fig.7 . in general , increasing the sweeping rate will reduce the probability of tunnelling to upper or lower level and the positive nonlinearity usually suppresses the probability of the state s staying in the mid - level , because that the nonlinearity with positive @xmath3 can be regarded as a kind of repulsive potential .
this repulsive self - interaction make particle tend to transition to lower level more easily , and this transition becomes more serious at the occurrence of the resonance between the `` internal field '' and external field .
the occurrence of the resonance is clearly exposed by the boundary between the white regime and dark regime in fig.7 . in the white regime , due to the resonance
, the nonlinearity dramatically changes the tunnelling probability .
the other issue we want to address is the symmetry .
the nonlinearity makes levels deform and therefore break the symmetry between upper level and lower level , consequently , the relations @xmath115 hold in linear case break in presence of the nonlinearity .
for our three - level system , the symmetry breaking is clearly exposed by fig.8 of showing the tunnelling probability @xmath12 as the functions of @xmath116 for @xmath39 , @xmath117 . in the linear case
, we have @xmath118 , however , with the presence of the nonlinear , @xmath119 increases whereas the @xmath120 decreases .
the similar things happen for the @xmath121 and @xmath122 .
the above symmetry breaking may be observed experimentally@xcite .
in conclusion , we have made a comprehensive analysis of the landau - zener tunnelling in a nonlinear three - level system , both analytically and numerically .
many novel tunnelling properties are demonstrated and behind dynamical mechanism is revealed . our model can be directly applied to the triple - well trapped bec and to explain the tunnelling dynamics between the traps @xcite . in a triple - trap @xmath123 ,
a bec is described by gross - pitaevskii equation ( gpe ) @xmath124 \psi ( r , t ) $ ] under the mean - field approximation , where @xmath125 , @xmath126 the atomic mass and @xmath127 the scattering length of the atom - atom interaction .
the wave function @xmath128 of gpe is the superposition of three wave functions describing the condensate in each trapraghavan , i.e. , @xmath129 .
when we study the tunnelling of three weakly coupled bec in traps @xmath130 , @xmath131 and @xmath132 , the dynamics of the system is described by the nonlinear schdinger equation with the hamiltonian , @xmath133where @xmath134 @xmath135 is the ground state energy for each trap . @xmath136
@xmath135 stands for atom - atom interaction , i.e. , nonlinear parameter .
@xmath137 is the coupling matrix element between trap @xmath130 and @xmath131 .
@xmath138 is the coupling matrix element between trap @xmath131 and @xmath132 . for simplicity
, we only consider the case that these two coupling matrix elements are the same and there is no coupling between trap @xmath130 and @xmath132 , i.e. , @xmath139 , @xmath140 .
the energy bias can be adjusted by tilting the trapping well and the nonlinearity can be adjusted by the feshbach resonance technique .
we hope our theory will stimulate the experiment in the direction .
this work was supported by national natural science foundation of china ( no.10474008,10445005 ) , science and technology fund of caep , the national fundamental research programme of china under grant no .
2005cb3724503 , the national high technology research and development program of china ( 863 program ) international cooperation program under grant no.2004aa1z1220 .
w. wernsdorfer , s. bhaduri , c. boskovic , g. christou , and d. n. hendrickson , phys .
b * 65 * , 180403 ( 2002 ) ; w. wernsdorfer , r. sessoli , a. caneschi , d. gatteschi , and a. cornia , europhys . lett . * 50 * , 552 ( 2000 ) . | we present a comprehensive analysis of the landau - zener tunnelling of a nonlinear three - level system in a linearly sweeping external field .
we find the presence of nonzero tunnelling probability in the adiabatic limit ( i.e. , very slowly sweeping field ) even for the situation that the nonlinear term is very small and the energy levels keep the same topological structure as that of linear case .
in particular , the tunnelling is irregular with showing an unresolved sensitivity on the sweeping rate . for the case of fast - sweeping fields
, we derive an analytic expression for the tunnelling probability with stationary phase approximation and show that the nonlinearity can dramatically influence the tunnelling probability when the nonlinear `` internal field '' resonate with the external field .
we also discuss the asymmetry of the tunnelling probability induced by the nonlinearity .
physics behind the above phenomena is revealed and possible application of our model to triple - well trapped bose - einstein condensate is discussed . |
quantum inversion has many applications in nuclear physics .
most of the potential descriptions of the few - body nuclear scattering base on the inversion of the scattering data .
though nucleon - nucleon interaction is used as input for all nuclear calculations , here we always have some fitted parameters and this fitting is , of course inversion .
moreover all modern high - precision nucleon - nucleon potentials are fitted to scattering data and are now perceived as phenomenology .
the problem here is `` that quantitative models for the nuclear force have only a poor theoretical background , while theory based models yield only poor results '' @xcite . at the same time the nucleon - nucleon phase shift analysis data are smooth in all investigated energy region up to 3 gev @xcite
this fact justifies the potential description without explicit internal degrees of freedom .
it is commonly supposed that nucleon - nucleon potential is nonlocal .
however local configuration space potentials simplify nuclear calculations greatly .
nonlocality effects may be treated as corrections due to the internal degrees of freedom . for these reasons construction of the high - precision local nn potential describing at least nn scattering data
is necessary for exact calculations in nuclear theory .
recently such results appeared in the literature @xcite . to simplify the following investigations we worked out the presented simple algorithm which allows to invert scattering data above inelasticity limit and to get the corresponding optical potential . in sect .
2 we describe our inversion algorithm , that allows to get the configuration space potential from phase shift analysis data neglecting inelasticity . in sect .
3 we show how to get an optical complex - valued potential from the real - valued one .
this optical potential describes phase shift analysis data and loss of flux due to inelastic processes . in sect .
4 we apply the worked out method to the nucleon - nucleon scattering data .
we extract two different optical potentials corresponding to the different asymptotic behavior of the scattering data .
the input data of the marchenko inversion are @xmath5 where @xmath6 is the scattering matrix dependance on the momentum @xmath7 , @xmath8 , @xmath9 is the energy of the j - th bound state , so that @xmath10 , @xmath11 is the particle ( reduced ) mass .
the @xmath12 matrices give the asymptotic behavior of the corresponding normalized bound states .
these are output data of the partial scattering schrdinger equation @xmath13\psi(r , q)=q^2\psi(r , q ) \label{schr}\ ] ] to illustrate the worked out algorithm first we consider the one channel case .
we proceed from the marchenko equation @xmath14 where @xmath15 @xmath16 are the riccati - hankel functions .
solution of the eq .
( [ eq2 ] ) is the function @xmath17 , which gives the reconstructed potential @xmath18 for the eq .
( [ schr ] ) @xmath19 contrary to the the algorithm used in @xcite we use the algebraic approximant not for the phase shifts @xmath20 but for the @xmath0-matrix @xmath21 or @xmath22 in the approximant ( [ eq5a ] ) @xmath23 and @xmath24 are an odd and even polynomials of @xmath7 , which do not turn to zero at the real axis simultaneously ( for @xmath25 , in the case then there is a bound state with @xmath26 , conversely @xmath23 and @xmath27 are even and odd polynomials ) . in the approximant ( [ eq5b ] ) @xmath28 and @xmath27 must be of different parity , but we can not describe the case with zero energy bound state .
the important feature of approximants ( [ eq5a ] ) and ( [ eq5b ] ) is that these are the most common pad approximants for the @xmath0-matrix consistent with its properties .
both approximants lead to the finite - dimensional kernel @xmath29 of the equation ( [ eq2 ] ) .
the function @xmath30 and potential @xmath31 are expressed through the elementary functions ( @xmath32 , @xmath33 and powers of @xmath34 ) , so we get bargmann potentials .
choice ( [ eq5a ] ) or ( [ eq5b ] ) is fixed for needed ( phenomenological ) dependance @xmath6 by numerical experiment .
we choose the relation which approximates @xmath6 better with less number of @xmath35 poles ( taking into account their multiplicity ) .
the less number of these poles gives the more simple potential .
the properties of the pad approximants are well known , and there are broad enough functional classes , for which this approximation converges everywhere besides poles of these functions . in particular , the consequence of the pad - hypothesis @xcite is that if a function is analytical at point @xmath36 and meromorphic at circle @xmath37 , containing this point , then its diagonal pad approximations @xmath38 $ ] ( where @xmath39 is the number of the approximation poles ) converges to this function on compact subsets of @xmath37 , which do not contain the poles of this function .
obviously for the major part of nuclear physics problems the @xmath0-matrix does satisfy these conditions . particularly for the short range
potentials the @xmath0-matrix is meromorohic at all complex plane .
another remark concerns the @xmath0-matrix poles positions .
our approximants for @xmath0-matrix ( [ eq5a ] ) , ( [ eq5b ] ) suggest that @xmath0-matrix poles may be positioned everywhere besides the real axis . previously at numerical approaches
the following @xmath0-matrix approximant was used @xcite : @xmath40 with real values @xmath41 , what assumes that @xmath0-matrix poles are positioned only at the imaginary axis .
it is obvious , our approximants are more common and allow to reconstruct the potentials of more complex form . the approximants ( [ eq5a ] ) and ( [ eq5b ] ) lead to the following expressions for the phase shifts @xmath42
@xmath43 or @xmath44 having the scattering data set @xmath45 ( @xmath46 , we get system of @xmath47 linear equations which define @xmath47 unknown coefficients of @xmath48 from ( [ eq6a ] ) and ( [ eq6b ] ) .
the number of experimental phase shifts differs generally from the chosen number @xmath47
. therefore the phenomenological dependance @xmath49 must be approximated by appropriate spline .
the @xmath47 needed values @xmath50 are defined by some selection from this spline .
the increase in number of precise experimental phase shifts would make approximants ( [ eq5a ] ) , ( [ eq5b ] ) more close to the true @xmath0-matrix .
the scattering data are defined experimentally and they have errors .
the arbitrary selection of @xmath51 ( @xmath46 from the experimentally defined allowable region may lead to close roots of the polynomials @xmath23 and @xmath27 , i.e. to false @xmath0-matrix poles .
therefore @xmath23 and @xmath27 must be factored and close roots particularly those situated near the real axis must be eliminated .
this elimination corresponds to elimination of the pad approximant defects .
furthermore , using the approximant ( [ eq5a ] ) ( in case of ( [ eq5b ] ) the solution is a particular case of the considered below solution for two bound channels ) and calculating the integral in the eq .
( [ eq3 ] ) using the residue theorem , we get the following expression for the kernel @xmath29 of eq .
( [ eq1 ] ) @xmath52 where @xmath53 ( @xmath54 are the @xmath0-matrix poles .
summing is taken over all the @xmath0-matrix poles @xmath55 positioned above the real axis and over bound states @xmath56 : @xmath57 .
so in this case the kernel of the eq .
( [ eq2 ] ) is finite - dimensional .
it is known that solution of such equations reduces to the linear equations solving .
we present the specific solution of this equation for our case .
we search the solution of eq .
( [ eq2 ] ) in the form @xmath58 where @xmath59 are unknown coefficients . substituting ( [ eq7 ] ) and ( [ eq8 ] ) into ( [ eq2 ] )
we get the following system of linear equations @xmath60 where @xmath61 the functional coefficients @xmath59 are defined by ( [ eq9 ] ) @xmath62 then @xmath30 and the potential @xmath63 from ( [ eq8 ] ) and ( [ eq4 ] ) .
the worked out technique of the inversion problem was approved on restoration of the square pit potentials .
the accuracy of the results is illustrated by initial and restored potentials in fig.1 .
the convergence of the restoration algorithm with increasing of power n of the @xmath64 } } } \right .
\kern-\nulldelimiterspace } { n]}$ ] pad approximant of the @xmath0-matrix is also shown here . in case of two bound channels
the system of the partial schrdinger equations is @xmath65 @xmath66 where @xmath67 , @xmath68 are potentials in channels 1 and 2 , @xmath69 is potential bounding them , @xmath70 and @xmath71 are channel wave functions .
by analogy with ( [ eq5b ] ) we approximate the @xmath0-matrix by the following expression @xmath72 here we again choose the most general form of pad approximant in contrast to the form used in @xcite .
the positions of @xmath0-matrix poles are determined from the equations analogous to ( [ eq6b ] ) , but with additional equations for the mixing parameter @xmath73 with scattering data @xmath74 . in case of bound channels and @xmath75 , in the original marchenko theory
it was proposed to use transformation of the initial eqs .
( [ eq10 ] ) with @xmath76 .
the same approach was used in @xcite .
nevertheless , from the numerical point of view this approach is less effective than the direct solution of the inverse problem with generalized marchenko equation @xcite .
formally it has the former view @xmath77 but functions involved are matrices @xmath78
@xmath79h\left ( { qy } \right)dq } + \sum\limits_{i = 1}^{n_{b } } { h\left ( { \beta _ i x } \right)m_i h\left ( { \beta _ i y } \right ) } , \ ] ] where @xmath80 using the chosen @xmath0-matrix approximant and the residue theorem we get the following expression for @xmath29 @xmath81 here @xmath82 are constant matrices , @xmath83 @xmath53 are the @xmath0-matrix poles , values @xmath84 correspond to bound states , @xmath85 is the set of @xmath0-matrix poles of the second order positioned above the real axis .
we solve eq .
( [ eq14 ] ) using substitution @xmath86 where summing is taken over all the @xmath0-matrix poles @xmath55 positioned above the real axis and over all values @xmath84 .
this construction leads to the system of linear equations for the functional @xmath87 matrix - coefficients @xmath59 , @xmath88 @xmath89 where @xmath90 @xmath91 @xmath92 @xmath93 @xmath94 is the unit matrix . the integrals in expressions ( [ eq19 ] )
are easily calculated analytically , but are cumbersome so we do not present them . having solved this linear equation system
we get the sought - for potential from ( [ eq18 ] ) and ( [ eq4 ] ) .
the multichannel generalization is made analogously .
the marchenko inversion does not give the needed optical potential . but found potential may serve as the initial potential for the iteration procedure that converts it into the optical ( complex - valued ) potential that describes inelastic processes .
first we consider the one channel problem .
the phase equation @xcite for the initial potential @xmath95 obtained by some inversion procedure ( say from marchenko equation ) is @xmath96 where @xmath97 is riccati - bessel amplitude @xcite .
let us consider the complex - valued potential @xmath98 obtained from @xmath95 by transformation @xmath99 where @xmath100 is some real parameter .
evidently the phase equation for this potential is @xmath101 from eqs .
( [ eq20 ] ) and ( [ eq22 ] ) we get @xmath102 for smooth enough potentials and @xmath103 not rapidly increasing with increasing of @xmath7 the right side of eq .
( [ eq23 ] ) rapidly decreases comparing with @xmath104 and @xmath105 , because under the integral in ( [ eq23 ] ) there is more rapidly oscillating function than in ( [ eq20 ] ) and ( [ eq22 ] ) .
then as a first approximation we may take @xmath106 for inelastic scattering the @xmath0-matrix is expressed through the real inelastic parameter @xmath107 and the real phase shift @xmath108 @xmath109 so it is easily arrived at @xmath110 @xmath111 whence it follows that @xmath112 .
the formula ( [ eq27 ] ) allows to calculate the parameter @xmath113 from the known values @xmath107 and @xmath114 .
this approximation works well enough for many quantum mechanical problems . for example in case of nucleon - nucleon scattering @xmath115 below the inelasticity limit which is high enough and @xmath107 grows slowly enough above this limit .
nevertheless in some instances the initial potential @xmath116 must be corrected . in this case
the following iteration procedure may be efficient .
first from the formula ( [ eq27 ] ) values @xmath117 are calculated from @xmath118 and @xmath20 . then with the potential @xmath119 , defined by formula ( [ eq21 ] ) new phase shifts @xmath120 are calculated from ( [ eq22 ] ) .
if they do not satisfy phase shift analysis data @xmath121 then changing the input data to @xmath122 we repeat inversion ( eqs .
( [ initial_for_mar])-([final_mar ] ) ) and determine new @xmath100 from formula ( [ eq27 ] ) .
the case of two channels is considered in a like manner , though the final expressions are more complex . by analogy with the one channel case
the following generalization for the optical potential is derived ( the mixing parameter @xmath123 is considered small ) : @xmath124 the first @xmath0-matrix approximation is @xmath125 as all the parameters and phase shifts are real , then eq . ( [ eq29 ] )
links between parameters @xmath100 and @xmath126 with the mixing parameters @xmath127 and phase shifts @xmath128 .
this consideration may be generalized to the multichannel case though calculations become much more complex .
we applied the described algorithm of inversion to reconstruction of the nucleon - nucleon potential . as input data for this reconstruction we used modern phase shift analysis data up to 1100 mev for @xmath129 state and up to 3 gev for @xmath130 state of nucleon - nucleon system @xcite .
the deuteron properties were taken from @xcite .
the relativistic effects were taken into account in frames of relativistic quantum mechanics of systems with a fixed number of particles ( point form dynamics ) .
the review of this approach can be found in @xcite .
this approach is based on the assumption that at not high energies we may consider the number of particles fixed , but the invariance group is the poincare group .
a system of two particles is described by the wave function , which is an eigenfunction of the mass operator or the mass squared operator . in this case
we may represent this wave function as a product of the external and internal wave functions @xcite .
the internal wave function is also an eigenfunction of the mass operator or the mass squared operator .
it is shown that the mass squared method is consistent with conventional fitting of the lorentz invariant cross section as a function of laboratory energy @xcite .
we consider system of two particles ( nucleons ) with equal masses .
then the internal wave function @xmath131 satisfies the following equation the eq .
( [ eq30b ] ) formally coincides with the schrdinger equation . in eqs .
( [ eq30 ] ) , ( [ eq30b ] ) @xmath135 is the mass squared operator , @xmath11 is the mass of nucleon , @xmath136 is the nucleon - nucleon potential , @xmath137 is the momentum operator of one of the nucleons in the center of masses system .
we use system @xmath138 .
the quasicoordinate representation corresponds to the realization @xmath139 , @xmath140 . in @xcite
we showed that this formalism can be easily generalized to the case of inelastic channels , particularly it allows to take into account the isobar channels in nn scattering .
this formal coincidence allows us to apply our inversion algorithm .
the modern phase shift analysis data @xcite allow to construct nucleon - nucleon potentials of two different kinds , depending on the asymptotic behavior of the phase shifts above investigated energy region of 3 gev .
we constructed the nucleon - nucleon optical potentials of two kinds for @xmath141 and @xmath142 partial waves .
these potentials describe the deuteron properties and the phase shift analysis data up to 3 gev and they have different behavior at short distances .
one is a repulsive core potential and another is a moscow attractive potential with forbidden states .
the moscow potential was introduced in @xcite .
these potentials are not phase equivalent and their @xmath0-matrices differ even bellow 3 gev , but within the experimental errors .
the @xmath141 phase shift of moscow potential begins from @xmath143 .
@xmath144 phase shifts of moscow potential begin from @xmath145 .
the mixing parameter @xmath146 of moscow potential differs from this of repulsive core potential by sign . above 3 gev
these potentials give different @xmath0-matrices .
the real parts of the constructed partial potentials are presented in fig . 2 and fig .
they may be downloaded from www.physics.khstu.ru in numerical form .
standard notation for central and tensor parts is used , so for @xmath142 the imaginary parts of potentials are defined by eqs .
( [ eq21 ] , [ eq28 ] ) , where parameters @xmath148 and @xmath149 can be easily calculated from the phase shift analysis data @xcite .
the phase shifts and mixing parameter are compared with the phase shift analysis data @xcite in fig .
4 . in table 1
the results of our deuteron properties calculations are compared with the experimental data @xcite . both kinds of the constructed partial potentials describe the data well in the limits of the experimental errors .
we conclude that the available data of the nn scattering and the deuteron properties are not discriminative with respect to the kind of the nn potential .
to discriminate between the considered two kinds of nucleon - nucleon potential we need careful experimental examination of other inelastic processes such as @xmath150 @xcite or @xmath151 .
arndt , i.i .
strakovsky and r.l .
workman , phys . rev .
* c62 * , 034005 ( 2000 ) ; r.a .
arndt , j.s .
hyslop and l.d .
roper , phys . rev . *
d35 * , 128 ( 1987 ) ; j bystricky , c. lechanoine - leluc , f. lehar , eur .
j. * c4 * , 607 ( 1998 ) ; j. ball , j. bystricky , j .-
fontaine et al , nuovo cimento * a111 * , 13 ( 1988 ) .
calculation + & & with repulsive & with moscow + & & core potential & potential + energy ( mev ) & 2,22458900(22 ) & 2,2246@xmath152 & 2,2246@xmath152 + @xmath153 ( fm@xmath154 ) & 0,2859(3 ) & 0,2639@xmath155 & 0,277@xmath155 + a@xmath156 ( fm@xmath157 & 0,8802(20 ) & 0,8802 & 0,8802 + r@xmath158 ( fm ) & 1,9627(38 ) & 1,951 & 1,956 + @xmath159 & 0,02714 & 0,02714 & 0,02714 + @xmath160 & 0,857406(1 ) & 0,8497@xmath155 & 0,859@xmath155 + | the numerical algorithm of the inverse quantum scattering is developed .
this algorithm is based on the marchenko theory , and includes three steps .
the first one is the algebraic pad approximation of the unitary s - matrix , what is realized by solving a system of linear equations .
second step is the exact solution of the marchenko equation .
the used approximant reduces it to another system of linear equations . at this step
we get the real - valued potential .
it is shown numerically that the developed algorithm is able not only to generate the given @xmath0-matrix dependence , but converges to the initial potential . at third step
we construct the optical complex - valued potential which gives the needed s - matrix .
it is shown that the modern phase shift analysis data allow to construct the nucleon - nucleon optical potentials of two kinds .
these potentials describe the deuteron properties and the phase shift analysis data up to 3 gev and they have different behavior at short distances .
one is a repulsive core potential and another is a moscow attractive potential with forbidden states .
* reconstruction of the optical potential in the inverse quantum scattering .
application to the relativistic inelastic nn scattering * n.a .
khokhlov@xmath1 , v.a .
knyr@xmath2 + @xmath3 _ e - mail : [email protected]_ ; @xmath4 _ e - mail : [email protected]_ |
as the most massive gravitationally bound systems in the universe , the rate of emergence of galaxy clusters since the big bang might be expected to be among the most straightforward predictions of cosmological models . yet despite the advent of the era of precision cosmology ushered in by observations of sne and the cosmic microwave background ( cmb ) , significant uncertainty remains in the expected numbers of galaxy clusters at @xmath0 .
the cmb temperature anisotropies on scales corresponding to clusters are not accurately known , leading to a range of values for @xmath11 , the rms matter fluctuation in a sphere of radius @xmath12 mpc at @xmath13 .
estimates of @xmath11 vary significantly , including e.g. @xmath14 @xcite , @xmath15 @xcite , @xmath16 @xcite , @xmath17 @xcite , @xmath18 @xcite , @xmath19 @xcite , and @xmath20 @xcite .
the range @xmath21 to 1 corresponds to a variation of a factor of nearly 20 in the predicted numbers of @xmath0 clusters with @xmath22 ( e.g. * ? ? ?
removing this uncertainty is a major goal of upcoming sunyaev - zeldovich cluster surveys such as the sza ( sunyaev - zeldovich array ; * ? ? ?
* ) , ami ( arcminute microkelvin imager ; * ? ? ?
* ) , act ( atacama cosmology telescope ; * ? ? ?
* ) , and spt ( south pole telescope ; * ? ? ?
since by definition , and for spectroscopically confirmed candidates in [ clusterz ] . ]
galaxy clusters contain an unusually high density of galaxies , they provide an efficient means of observing substantial numbers of galaxies at a common distance , offering the hope of constructing the analog of the hertzsprung - russell ( i.e. color - magnitude ) diagram for galaxy evolution .
indeed , studies of the relationship between color and magnitude indicate that clusters are the habitat of galaxies with the oldest and most massive stellar populations ( e.g. @xcite ; @xcite ) , objects reasonably free of the complications associated with starbursts and dust .
these studies are consistent with an extremely simple formation history for cluster galaxies , in which their stars are formed in a short burst at high redshift , and they evolve quiescently thereafter ( we use the term `` red spike model '' in referring to this scenario - see figure [ m_vs_z ] ) .
studies of the near - ir luminosity functions of cluster galaxies reinforce this picture ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) . with their large lookback times ,
therefore , high redshift galaxy clusters also provide an observational pillar for our understanding of the formation and evolution of galaxies . obtaining substantial samples of galaxy clusters at @xmath0
has proved challenging , largely because such objects are difficult to detect using only optical data .
due to their greatly enhanced rate of star formation by @xmath8 , the uv emission from modest sized field galaxies overwhelms that from the intrinsically red spectra of quiesecent , early type galaxies preferentially found in clusters .
the red sequence cluster survey @xcite uses the observed color - magnitude relationship in cluster galaxies to improve the contrast and has proven highly efficient to @xmath8 , but the optical colors of the red sequence become increasingly degenerate at higher redshifts , as they no longer span the rest 4000 break .
@xcite describe a program to extend the red sequence technique to higher redshift using spitzer data , but it is also important to _ test _ for the existence of red sequences in @xmath0 clusters rather than preselecting for them , if possible .
the contrast of high redshift clusters over the field improves at longer wavelengths ( figure [ m_vs_z ] ) , but the contrast against atmospheric emission declines , and until recently the relatively small formats of infrared detector arrays made surveying sufficient @xmath23 volume a formidable undertaking .
@xcite reported the discovery of a cluster at @xmath24 in a 100 square arcminute survey to @xmath25(vega ) = 20 ( 10@xmath26 ) . but this survey required approximately 2 hours of exposure in both @xmath27 and @xmath25 per position , and 30 allocated nights of kpno 4 m time to complete . with estimates for the surface density of @xmath6 clusters at @xmath0 in the range @xmath28 per square degree @xcite ,
the discovery was in hindsight fortuitous .
such considerations motivated a different approach , where extended sources in deep x - ray surveys lacking prominent optical counterparts were targeted for ir followup .
this technique yielded confirmed clusters at @xmath29 1.23 , and 1.26 @xcite . with the arrival of _ xmm _ ,
x - ray surveys offer renewed promise , leading recently to the identification of galaxy clusters at @xmath30 @xcite and 1.45 @xcite . with exposure times @xmath31 ksec and a 30 arcmin field of view , a discovery rate of approximately 30 hours per candidate @xmath0 cluster above @xmath6
is expected ( assuming one such cluster per square degree , which corresponds to @xmath32 ) .
searches for clusters around radio galaxies have yielded protoclusters with redshifts as high as 4.1 @xcite and possibly even 5.2 @xcite .
the very large redshifts of these systems enable powerful inferences to be drawn regarding the formation of cluster galaxies , but they are less useful as probes of the cosmological growth of structure .
lyman break galaxy surveys with intensive followup spectroscopy on the keck telescopes have also identified highly overdense structures at @xmath33 and 3.09 @xcite , and @xcite discuss a @xmath34 structure identified via ly-@xmath35 emission in a narrow band imaging survey .
recent advancements in ir detector array formats have renewed interest in ground - based ir surveys . in one example of the state of the art , elston , gonzalez et al .
( 2006 ) use the @xmath36 pixel flamingos camera to map 4 deg@xmath2 to a 50% completeness limit of @xmath37 ( vega ) . with 2 hour exposures on the kpno 2.1
m each covering 1/10th deg@xmath2 , this leads to an expected discovery rate for high redshift clusters per useful hour of observing which is similar to _
xmm_. the ukidss ultra deep survey provides another recent example , finding 13 cluster candidates with @xmath38 in a 0.5 deg@xmath2 survey @xcite , one of which has 4 spectroscopic redshifts at @xmath39 @xcite , and @xcite find 12 candidates with @xmath40 in a 0.66 deg@xmath2 @xmath41-band survey in the cosmos field .
@xcite present a system with a high density of galaxies with red optical to near - ir colors surrounding a galaxy at @xmath42 , identified in the 120 arcmin@xmath2 gemini deep deep survey .
candidates drawn from surveys of less than a square degree are unlikely to include many rich clusters , however . with the launch of the _ spitzer space telescope _ in 2003 @xcite ,
sensitive infrared arrays free from foreground thermal emission were put into operation @xcite . a major scientific driver for the _ spitzer _
infrared array camera ( irac ) shallow survey @xcite was the detection of @xmath0 galaxy clusters .
the irac shallow survey uses 90 second exposures per position and covers 8.5 deg@xmath2 , leading to an expected discovery rate of @xmath43 hours per @xmath0 cluster . here
we present results from the irac shallow survey cluster search , finding 106 cluster and group candidates at @xmath0 , of which we estimate only @xmath3 are spurious .
a surface density of over 10 systems per square degree at @xmath0 is higher than expected for bound systems with masses above @xmath6 for the current range of plausible @xmath11 estimates . while we present evidence that at least two of the @xmath0 clusters have masses well above @xmath6 , it is likely that our sample includes systems with masses below @xmath6 ( i.e. groups ) , and perhaps some unbound filaments viewed end - on . in the remainder of this paper , for brevity the terms clusters " and candidates " are used to refer to all such objects which meet our selection criteria , unless otherwise stated . this paper describes how the cluster sample was identified , and some of the overall photometric properties of the sample .
we also provide new spectroscopic evidence supporting nine of these clusters , from @xmath44 to 1.373 .
spectroscopic evidence in support of irac shallow survey selected clusters at @xmath45 1.243 , and 1.413 was presented in @xcite , @xcite , and @xcite respectively , and we provide additional previously unpublished spectroscopy on those clusters here , for completeness . @xcite
discuss the clustering of the clusters , and galametz et al .
( in preparation ) report on agn incidence vs cluster - centric distance
. followup imaging with _ hst _
( go 10496 , perlmutter ; 10836 , stanford ; and 11002 , eisenhardt ) and _ spitzer _ ( go 30950 , eisenhardt ) is underway , and future papers will examine the scatter in the color - magnitude relation as a function of morphological type , the dependence of cluster galaxy size on redshift , starburst activity in clusters vs. redshift , and the dependence of mean galaxy properties on surface density . a cosmology with @xmath46 km s@xmath5 , @xmath47 , @xmath48 is assumed , and magnitudes are on the vega system ( defined in * ? ? ? * for irac ) . at @xmath49 ,
this means that one arcminute corresponds to a physical scale of @xmath50 kpc , peaking at 508 kpc at @xmath51 .
unless otherwise specified , physical ( rather than co - moving ) scales are used throughout .
the irac shallow survey @xcite was designed to maximize the number of reliable sources detected per unit time and to cover sufficient area to detect significant numbers of @xmath23 galaxy clusters . as explained in @xcite ,
a 30 second exposure time per pointing is close to optimum for maximizing source detections , and reliability was obtained by requiring three independent exposures separated by hours at each position .
the survey covers @xmath52 8.5 deg@xmath2 and reaches an aperture - corrected 5@xmath26 depth of @xmath52 19.1 and 18.3 mag ( vega ) at 3.6 and @xmath53 m in 3 diameter apertures .
it is a remarkable fact that 90 seconds of combined exposure with irac on the 85 cm _
spitzer space telescope _
provides sufficient sensitivity to detect evolving @xmath54 galaxies to @xmath55 ( figure 1 ) .
the survey was carried out in the botes region of the noao deep wide - field survey ( ndwfs ; * ? ? ?
* ) to allow photometric redshifts to be derived using the deep optical imaging available for this field .
the ndwfs reaches 5@xmath26 point - source depths in the @xmath56 and @xmath57 bands of @xmath52 27.1 , 26.1 , and 25.4 respectively .
typical exposure times per position with the mosaic-1 camera on the kpno mayall 4-m telescope were 1 2 hours in @xmath58 , 1 2 hours in @xmath59 , and 2 4 hours in @xmath57 , and the seeing ranged from 0.7 to 1.5 . the data acquisition , reduction , and catalog generation are discussed in detail by b. jannuzi et al .
( in preparation ) and a. dey et al .
( in preparation ) .
this paper uses the ndwfs third data release ( dr3 ) images and sextractor catalogs which can be obtained through the noao data archive . the agn and galaxy evolution survey ( ages , c. kochanek et al . in preparation ) provides spectroscopic redshifts for @xmath60 objects ( using the version 2.0 catalog ) in the irac shallow survey .
ages is highly complete for sources brighter than 15.7 mag at @xmath53 m , and also for sources brighter than @xmath61 , with many redshifts for sources up to @xmath62 , enabling excellent assessment of the photometric redshifts to @xmath63 ( figure 1 ) .
deep near infrared imaging from the flamingos extragalactic survey ( flamex ) is available for half the botes region @xcite , and was used for deriving a prior on the redshift likelihood functions ( see [ sec : photz ] ) .
imaging of the botes ndwfs field has also been obtained in the radio @xcite , at 24 , 70 , and @xmath64 m with the mips instrument on _ spitzer _
@xcite , in the @xmath65band @xcite , in the @xmath66 with @xmath67 , and in x - rays to a depth of 5 ksec with the acis instrument on the _ chandra x - ray observatory _ @xcite , but these data are not used in this paper .
object selection for the cluster search was carried out in the @xmath1 m band , because the negative k - correction as the rest frame @xmath68 m peak shifts into this band leads to a flux which is nearly independent of redshift for @xmath69 ( figure 1 ) .
object detection and photometry was carried out using sextractor @xcite in double image mode , allowing matched aperture photometry in the other irac bands .
while smaller apertures maximize the depth of the survey , monte carlo simulations showed that 5 diameter aperture magnitudes are necessary to provide sufficiently reliable color measurements and photometric redshifts @xcite . this flux limit ( 5@xmath26 at @xmath53 m in 5 )
corresponds to @xmath70 @xmath71jy , or a vega based magnitude of @xmath72 .
the ndwfs catalogs were also generated using sextractor , but run in single - image mode in each band .
detections in the different optical bands and between the optical and irac catalogs were matched if the centroids were within @xmath73 of each other , using the closest optical source if more than one satisfied this criterion .
for very extended objects ( generally at @xmath74 ) , detections in the different bands were matched if the centroids were within an ellipse defined using the second order moments of the light distribution of the object @xcite .
the ndwfs data were taken over several years in variable conditions , and therefore the photometric depths vary somewhat from pointing to pointing .
average 50% completeness limits for the @xmath58 , @xmath59 , and @xmath57bands were 26.7 , 25.6 , and 25.0 respectively .
average 5@xmath26 flux ( magnitude ) limits in a 5aperture ( which is significantly larger than the optimum detection aperture for these data ) were measured via monte carlo simulations to be @xmath76 @xmath71jy ( @xmath77 mag ) in @xmath58 , @xmath78 @xmath71jy ( @xmath79 mag ) in @xmath59 , and @xmath80 @xmath71jy ( @xmath81 mag ) in the @xmath57band @xcite . in the space based irac shallow survey the depth is more uniform , and in the @xmath82 m band it is @xmath83 @xmath71jy ( @xmath84 mag ) , also derived from a monte carlo simulation .
although the majority of objects in the @xmath1m - selected catalog are well detected at shorter wavelengths , this is not generally the case for @xmath85 red ellipticals .
objects at these redshifts are often quite faint in the optical as the 4000 break is longward of the @xmath57band ( see figure 1 ) . where sources were observed but not detected , the flux was taken to be zero and a monte carlo 1@xmath26 error was adopted .
this approach is optimal for photometric redshift fitting , where the non - detection provides important contraints on the galaxy spectral energy distribution ( sed ) .
the photometry used for photometric redshift estimation ( [ sec : photz ] ) consists of @xmath86[4.5]$ ] data with the monte carlo photometric errors and limits noted above .
the @xmath87 m and @xmath88 m bands were not used because they do not have the sensitivity to detect @xmath85 cluster @xmath54 galaxies in the irac shallow survey .
while the irac shallow survey covers 8.5 @xmath89 in each band , the overlap area observed in both the 3.6 and @xmath53 m irac bands is 8.0 @xmath89 .
all of this falls within the optical ndwfs botes region .
however , due to haloes around bright objects , shorter observation times for regions at the edges of individual mosaic camera pointings , and residual ccd defects , 0.75 @xmath89 may have lower quality optical photometry and hence photometric redshifts . hence these regions are also excluded from the sample .
finally , 14,044 stars were removed from the catalog using the sextractor stellarity index in the best seeing optical data for each location .
comparison to the star count model of @xcite , as tabulated in @xcite , suggests that @xmath90 of stars were identified via this approach , leaving @xmath91 of the @xmath1 m sample as unrecognized stars . in summary ,
photometric redshifts were estimated for a total of 175,431 objects brighter than @xmath70 @xmath71jy ( @xmath92 ) at @xmath53 m in 5 in a 7.25 @xmath89 region .
a full description of the photometric redshift methodology is given in @xcite .
a summary of those aspects most relevant to cluster detection is provided here .
photometric redshifts were computed using an empirical template
fitting algorithm which linearly interpolates between the four coleman , wu , & weedman ( 1980 ) seds ( e , sbc , scd , and i m ) , augmented by the @xcite sb3 and sb2 starburst templates .
these seds were extended to the far uv and near ir using @xcite models .
these stellar photospheric models do not include emission from dust , in particular the pah features which dominate the @xmath93 m portion of the spectrum in starforming galaxies , but at @xmath0 the photometry used does not sample these rest wavelengths .
in addition to the large ages ( c. kochanek et al . in preparation ) survey , there are @xmath94 spectroscopic redshifts extending to @xmath95 gleaned from several ongoing surveys in the botes field .
these were used as training sets to adjust the templates and photometric zero points to improve overall redshift accuracy and reliability ( see * ? ? ? * for details ) .
comparison with these spectroscopic samples shows that an rms dispersion of @xmath96 is achieved for 95% of galaxies to at least @xmath97 .
subsequent follow up spectroscopy of high redshift candidate clusters ( [ sec : keck ] ; see also * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) verify this accuracy .
a key output of the @xcite technique is the redshift probability function for each object , @xmath98 , derived directly from the redshift - axis projection of the full redshift sed likelihood surface . to transform these simple redshift likelihood functions into true probability distribution functions ,
a prior consisting of the observed redshift distribution was applied .
this was measured in the flamex region using high
quality @xmath99[4.5]$ ] photometric redshifts .
comparison to the spectroscopic sample illustrates that the resulting distribution functions are statistically valid in the sense that integrated areas accurately represent redshift probabilities at the 1 , 2 , and 3@xmath26 levels .
these @xmath98 functions are the input to the wavelet detection algorithm discussed in
[ sec : detn ] and are also used in [ sec : discussion ] .
note that the @xmath98 distributions shows relatively little dependence on galaxy type @xcite .
the reason is that while all galaxies are selected to have at least 5 sigma detections in [ 4.5 ] , red galaxies are often only marginally detected or even undetected in our optical images , since they have very little blue light .
this is particularly true at @xmath85 where the @xmath100-correction for early types is large in the optical .
blue galaxies have good @xmath101[3.6][4.5 ] photometry , because they have more blue light .
thus even though blue , late - type galaxy sed s have smaller breaks , they have more extensive ( useful ) photometry . these compensating effects lead to effectively type - independent photometric redshifts for @xmath85 galaxies .
we employed a wavelet analysis to identify galaxy clusters within the botes region .
wavelet decomposition is a commonly used technique for cluster identification in x - ray images ( for example , see * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , where it provides an effective means of identifying extended sources in the presence of contaminating point sources . in principle
, a similar analysis can be used with optical and infrared data sets , using galaxies rather than x - ray photons to identify extended sources . as is well known ,
galaxy number counts are more susceptible to projection effects than is bremsstrahlung emission from the icm .
this issue is one that must be dealt with for all optical and infrared cluster searches , and consequently most such searches for distant clusters make explicit assumptions about the properties of the distant cluster population , such as an assumed density profile ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) or the presence of a red sequence @xcite .
the sdss c4 technique @xcite does not require a _ red _ sequence , but it does demand that the colors of cluster galaxies are similar to one another .
a significant advantage of the botes data set is that the photometric redshifts permit such assumptions to be minimized .
the full photometric redshift probability distributions @xmath98 were used to construct weighted galaxy density maps within overlapping redshift slices of width @xmath102 , stepping through redshift space in increments of @xmath103 . for each galaxy
the weight in the map corresponds to the probability that the galaxy lies within the given redshift slice .
weighting in this fashion de - emphasizes sources for which the redshift is poorly constrained .
it is worth emphasizing that _ all _ galaxies were included in construction of the density map , and consequently cluster detection should be relatively independent of sed type and hence independent of morphology .
finally , cluster detection will be insensitive to the resolution of the density maps provided that the pixel size is small compared to the angular extent of cluster cores at all redshifts .
a resolution of 12@xmath104 kpc ) per pixel was used , which satisfies this criterion while being sufficiently large to keep computational overhead manageable .
galaxy cluster candidates were detected within each redshift slice by convolving the density map with the wavelet kernel .
we use a gaussian difference kernel of the form @xmath105 where @xmath106 kpc and @xmath107 kpc , and which crosses zero near @xmath108 mpc .
the scale of the kernel is fixed in physical rather than angular units , preserving our ability to uniformly identify comparable systems at different redshifts .
the precise physical values of @xmath26 are subject to refinement , but the selected values effectively isolate overdensities on the scale of clusters or groups . galaxy cluster candidates were detected in each redshift slice of these wavelet smoothed galaxy density maps using a simple peak - finding algorithm . to establish a consistent significance level for the candidates , 1000 bootstrap simulations were carried out within each redshift slice .
the existing p(z ) distributions , right ascensions , and declinations were repeatedly shuffled , convolved with the wavelet kernel , and candidates detected to find the threshold corresponding to one false positive per redshift slice within the botes field .
a list of detections above this significance threshold was generated for each redshift slice . because the right ascensions and declinations were shuffled independently , the true correlated background was not preserved .
consequently , the contamination rate may be somewhat higher than the one false positive per redshift slice expected if we had preserved the correlated background .
for the current analysis we accept the somewhat higher contamination rate in exchange for improved completeness , particularly for the highest redshift clusters .
the majority of clusters are detected in multiple redshift slices ( a natural consequence of sampling in slices separated by step sizes finer than the galaxy redshift uncertainties ) .
multiple detections with small separations in positions and redshift slices were considered to be a single cluster , which was assigned an estimated redshift and position corresponding to the slice with the highest statistical significance .
a total of 335 candidates were found in redshift slices with centers from 0.1 to 1.9 , including 98 in slices with @xmath85 . while the detection technique is on three dimensional overdensities , it is not immune to projection effects .
the expected number of clusters in each cylindrical bin of length @xmath102 and radius 1 mpc ( the radius of the wavelet detection kernel ) was computed to quantify the extent of projection .
this calculation included both the random expectation due to the observed number density and the excess clusters expected due to the observed clustering of this sample @xcite .
the projection rate is most significant for lower redshifts because of the large angular size corresponding to 1 mpc , affecting @xmath109 of systems at @xmath110 to @xmath111 at @xmath112 .
the projection rate decreases to below @xmath113% at @xmath114 and is negligible at @xmath115 .
overall we estimate that @xmath3 of the candidates may be spurious , including the contamination noted in the previous paragraph .
the peak of the summed @xmath98 distribution at the cluster detection location was taken as the initial estimate of each cluster s redshift . to improve this estimate , individual objects within a 1 mpc radius of the cluster which included the cluster redshift within the @xmath116 range of their @xmath98 functions
were considered candidate cluster members . a refined estimate ( @xmath117 ) of the mean cluster redshift
was obtained from the peak of the summed @xmath98 distribution for these members . from this analysis ,
104 candidates have @xmath118 . in table [ z1table ] and figures 2 - 13 we present 12 of the cluster candidates that were detected via the above criteria , and subsequently spectroscopically confirmed at the keck observatory to lie at @xmath85 ( see [ sec : keck ] ) .
two of these , iscs j1434.1 + 3328 and iscs j1429.2 + 3357 , have @xmath119 but the spectroscopic mean redshifts are @xmath44 and @xmath120 respectively .
hence we report a total of 106 clusters at @xmath0 , of which roughly 10% may be expected to arise by chance or from projection effects , for the reasons noted in [ sec : detn ] .
column 1 of table 1 provides the catalog number of each cluster .
the catalog numbers increase with decreasing detection significance ( [ sec : detn ] ) .
column 2 is the iau designation for each cluster , based on the ( j2000 ) coordinates of the detection given in columns 3 and 4 .
column 5 provides the @xmath117 value described above , and column 6 gives the mean redshift of the spectroscopically confirmed members .
column 7 provides the number of photometric redshift members of the cluster , defined as galaxies within 1 mpc of the cluster center and with integrated @xmath121 in the range @xmath122 .
these galaxies are used to calculate mean cluster colors in [ sec : im3p6_vs_z ] .
column 8 reduces the number in column 7 by the number of galaxies which satisfy these criteria for each cluster redshift over the entire field , scaled to the 1 mpc radius area for each cluster redshift .
column 9 gives the number of spectroscopically confirmed member galaxies in each cluster ( see below ) .
column 10 gives the mean @xmath7 $ ] color for the photometric redshift member galaxies , and column 11 provides the sum of their luminosities in [ 4.5 ] relative to an @xmath54 value which evolves according to the `` red spike '' model shown in figure [ m_vs_z ] , corrected for the average over the field .
column 12 gives the luminosity in [ 4.5 ] of the brightest photometric redshift member galaxy relative to @xmath54 .
we define a @xmath0 cluster as spectroscopically confirmed if it contains at least 5 galaxies in the range @xmath122 and within a radius of 2 mpc , whose spectroscopic redshifts match to within @xmath123 km / s
. the spectroscopic redshifts must also be of class a or b. class a spectra have unambiguous redshift determinations , typically relying upon multiple well detected emission or absorption features .
class b spectral features are reliable but are less well detected .
the radius threshold for spectroscopic membership is larger than the 1 mpc used for photometric redshift members for practical reasons : most of the @xmath0 redshifts reported here were obtained with slitmasks extending out to approximately 2 mpc from the cluster center .
typically @xmath124 photometric redshift members within 1 mpc could be accommodated in the mask design , and often no redshift could be determined from the resulting spectra .
the least significant cluster in table 1 is iscs j1434.5 + 3427 , which is the 327th most significant detection out of the sample of 335 , and the 100th most significant detection at @xmath0 . despite its relatively low detection significance , iscs j1434.5 + 3427
has a striking filamentary morphology ( figure [ color10.342 ] ) , and has eleven spectroscopically confirmed members ( see also * ? ? ?
lcrccccccccc 152 & iscs_j1434.1 + 3328 & 14:34:10.37 & + 33:28:18.3 & 0.98 & 1.057 & 32 & 13 & 6 & 4.83 & 15 & 2.5 + 51 & iscs_j1429.2 + 3357 & 14:29:15.16 & + 33:57:08.5 & 0.98 & 1.058 & 45 & 26 & 7 & 4.80 & 34 & 4.2 + 19 & iscs_j1433.1 + 3334 & 14:33:06.81 & + 33:34:14.2 & 1.02 & 1.070 & 57 & 38 & 20 & 4.93 & 45 & 4.3 + 123 & iscs_j1433.2 + 3324 & 14:33:16.01 & + 33:24:37.4 & 1.01 & 1.096 & 31 & 11 & 6 & 4.94 & 15 & 4.6 + 17 & iscs_j1432.4 + 3332 & 14:32:29.18 & + 33:32:36.0 & 1.08 & 1.112 & 49 & 31 & 23 & 5.16 & 47 & 5.3 + 34 & iscs_j1426.1 + 3403 & 14:26:09.51 & + 34:03:41.1 & 1.08 & 1.135 & 31 & 13 & 7 & 5.18 & 26 & 4.0 + 14 & iscs_j1426.5 + 3339 & 14:26:30.42 & + 33:39:33.2 & 1.11 & 1.161 & 52 & 35 & 5 & 5.15 & 47 & 2.8 + 342 & iscs_j1434.5 + 3427 & 14:34:30.44 & + 34:27:12.3 & 1.20 & 1.243 & 27 & 12 & 11 & 5.51 & 27 & 6.0 + 30 & iscs_j1429.3 + 3437 & 14:29:18.51 & + 34:37:25.8 & 1.14 & 1.258 & 23 & 7 & 9 & 5.30 & 11 & 6.5 + 29 & iscs_j1432.6 + 3436 & 14:32:38.38 & + 34:36:49.0 & 1.24 & 1.347 & 30 & 17 & 8 & 5.65 & 31 & 4.3 + 25 & iscs_j1434.7 + 3519 & 14:34:46.33 & + 35:19:33.5 & 1.37 & 1.373 & 19 & 9 & 5 & 5.77 & 17 & 3.9 + 22 & iscs_j1438.1 + 3414 & 14:38:08.71 & + 34:14:19.2 & 1.33 & 1.413 & 25 & 15 & 10 & 5.75 & 24 & 3.5 + the extensive ages spectroscopic database ( c. kochanek et al . in preparation ) can be used to spectroscopically confirm clusters at @xmath125 , a redshift at which @xmath54 corresponds to @xmath126 ( figure 1 ) , the magnitude limit for ages spectroscopy .
note that while ages is 94% complete to @xmath127 , sparse sampling is used for fainter galaxies , and currently spectra are available for @xmath128 of galaxies with @xmath129 . because of the high surface density of ages redshifts and the larger angular scale at @xmath125 , and the fact that the [ 4.5 ] flux limit samples substantially less luminous galaxies at these redshifts , two additions to the criteria for spectroscopic confirmation used at @xmath0 were imposed . to contribute to spectroscopic confirmation for cluster candidates with @xmath130
, galaxies were required to be more luminous than l * + 1 in [ 4.5 ] for a red spike evolving model .
also , the average surface density of ages galaxies over the botes field which would satisfy our high redshift confirmation criteria was calculated as a function of redshift , and 5 galaxies above this field level were required for a cluster to be confirmed . of the 335 cluster candidates ,
80 have @xmath130 .
however seven lie near the edges of the botes field and hence were not well observed in ages . of the remaining 73 , 61 candidates are confirmed by ages spectroscopy using the criteria just described .
this criterion is perhaps overly stringent : it rejects two clusters at @xmath131 with 17 or more matching ages redshifts , but with only 4 ( rather than 5 ) for galaxies brighter than l * + 1 after field correction .
all of the other 10 candidates observed by ages which do not meet the confirmation criteria have @xmath132 , and half have @xmath133 . given the sparse sampling of ages at @xmath134 , which corresponds to @xmath54 at @xmath135 ( figure 1 ) , we believe this confirmation rate validates our estimate that only @xmath109 of our cluster candidates arise by chance or due to projection effects .
further details on @xmath136 clusters will be provided in a. gonzalez et al .
( in preparation )
. most of the high redshift spectroscopic confirmation of iscs clusters has been obtained at keck observatory .
three clusters observed with keck have been reported previously : iscs j1432.4 + 3332 ( @xmath137 ) , iscs j1434.5 + 3427 ( @xmath138 ) , and iscs j1438.1 + 3414 ( @xmath139 ) are presented in @xcite , @xcite , and @xcite , respectively . here
we provide spectroscopic confirmation for an additional nine clusters at @xmath0 , as well as some new spectroscopic information for the initial three clusters .
table [ specobstable ] details new observations of iscs clusters .
table [ spectable ] provides properties of previously unreported , spectroscopically confirmed cluster members , in declination order for irac sources , followed by serendipitous sources .
figure [ spectra ] shows three example spectra obtained in april 2007 with keck / deimos .
lcrcccccccc iscs_j1434.1 + 3328 & 1.057 & deimos & 2007 apr 18@xmath14019 & 6@xmath1411800 & clear ; 08 - 20 + iscs_j1429.2 + 3357 & 1.058 & deimos & 2006 apr 26 & 4@xmath1411500 & not photometric + iscs_j1433.1 + 3334 & 1.070 & deimos & 2007 apr 19 & 3@xmath1411200 & clear ; 08 + iscs_j1433.2 + 3324 & 1.096 & deimos & 2007 apr 18@xmath14019 & 6@xmath1411800 & clear ; 08 - 20 + iscs_j1432.4 + 3332 & 1.112 & lris & 2005 jun 03 & 5@xmath1411800 & clear ; 09 ( elston et al .
2006 ) + & & focas & 2006 apr 21 & 5@xmath1411200 & + & & deimos & 2007 apr 19 & 3@xmath1411200 & clear ; 08 + iscs_j1426.1 + 3403 & 1.135 & lris & 2007 may 19 & 3@xmath1411800 & clear ; 09 + iscs_j1426.5 + 3339 & 1.161 & lris & 2006 apr 04 & 4@xmath1411800 & clear + iscs_j1434.5 + 3427 & 1.243 & focas & 2006 apr 22 & 5@xmath1411200 & + & & deimos & 2007 apr 18 & 8@xmath1411800 & likely cirrus ; 10 - 15 + iscs_j1429.3 + 3437 & 1.258 & lris & 2006 apr 05 & 3@xmath1411200 & + iscs_j1432.6 + 3436 & 1.347 & lris & 2007 may 21 & 7@xmath1411800 & mostly clear ; 08 + iscs_j1434.7 + 3519 & 1.373 & lris & 2005 jun 02 & 7@xmath1411800 & + iscs_j1438.1 + 3414 & 1.413 & focas & 2006 jun 28 & 9@xmath1411200 & + & & deimos & 2007 apr 19 & 7@xmath1411800 & clear ; 09
+ we obtained deep optical slitmask spectroscopy for several clusters using the dual - beam low resolution imaging spectrograph ( lris ; * ? ? ? * ) on the 10 m keck i telescope during three observing runs between 2005 and 2007 .
slitmasks generally included approximately 15 objects with photometric redshifts consistent with cluster membership and within 4 arcmin of the nominal cluster centers .
additional irac 4.5@xmath71 m selected sources were included to fill out the slitmasks .
slitlets had widths of 1.3 arcsec and minimum lengths of 10 arcsec .
we employed the d580 dichroic which splits the light at @xmath142 between the two channels of lris . on the red side , the 400 line grating , blazed at 8500 ,
was used to cover a nominal wavelength range of 5800 to 9800 , varying somewhat depending on the position of a slit in the mask . for objects filling a slitlet ,
the spectral resolution for this instrument configuration is @xmath143 ( @xmath144 ) , as determined from arc lamp spectra . on the blue side ,
the 400 line grism , blazed at 3400 , provided coverage from the atmospheric cutoff ( @xmath145 ) up to the dichroic cut off . for objects filling a slitlet ,
the spectral resolution for this instrument configuration is @xmath146 ( @xmath147 ) .
we obtained multiple exposures for each mask , usually with 1800 s per individual exposure .
table [ specobstable ] details the exposure times and observing conditions .
the observations were carried out with the slitlets aligned close to the parallactic angle , and objects were shifted along the long axis of the slitlets between exposures to enable better sky subtraction and fringe correction .
the slitmask data were separated into individual spectra and then reduced using standard longslit techniques .
the multiple exposures for each slitlet were reduced separately and then coadded . the spectra were reduced both without and with a fringe correction ; the former tends to yield higher quality object spectra at the shorter wavelengths , while the latter is necessary at the longer wavelengths .
calibrations were obtained from arc lamp exposures taken immediately after the object exposures for the red side , and from arc lamp exposures taken during the afternoon for the blue side .
corrections for small offsets in the wavelength calibration were obtained by inspection of the positions of sky lines in the object spectra .
using longslit observations of the standard stars from @xcite obtained during the same observing runs , we achieved relative flux calibration of the spectroscopy . while slit losses for resolved sources preclude absolute spectrophotometry from the slit mask data , the relative calibration of the spectral shapes should be accurate .
one dimensional spectra were extracted from the sum of all the reduced data for each slitlet for both the red and blue sides . for the targets in the high - redshift clusters , generally only the red side data proved useful .
additional spectroscopy was obtained with the deep imaging multi - object spectrograph ( deimos ; * ? ? ?
* ) on the 10 m keck ii telescope , a second generation instrument with significantly more multiplexing capabilities as compared to lris , albeit without the blue sensitivity .
during an observing run in april 2006 , we targeted botes active galaxy candidates selected on the basis of mid - infrared colors ( e.g. , * ? ? ? * ) , but included candidate cluster members in one mask . in april 2007 , while observing host galaxies of high redshift cluster supernovae ( perlmutter et al .
, in preparation ) , we also observed candidate cluster members , typically with more than one high redshift cluster candidate observed in each wide area mask . for both observing runs , the 600zd grating ( @xmath148 ; @xmath149 ) and a gg455 order - blocking filter were used .
deimos data were processed using a slightly modified version of the pipeline developed by the deep2 team at uc - berkeley .
although neither run was completely photometric , relative flux calibration was achieved from observations of standard stars from @xcite .
followup spectroscopic observations were also obtained with the focas spectrograph @xcite on the 8.2-m subaru telescope in april and june 2006 .
for these observations , we used the instrument in multi - object spectroscopy mode with 0.8 arcsec width slits , 300r grism , and so58 order - sorting filter , which provided about 15 spectra from 5800 to 10000 with a spectral resolution @xmath150 .
total exposure times were 2 - 6 hours .
the focas data were reduced with iraf using standard methods including a fringe correction .
wavelength calibration was done using oh airglow emission lines .
absolute flux was calibrated using standard star ( feige 34 , wolf 1346 , hz 44 ) spectra taken during the same nights .
iscs j1434.1 + 3328 was observed with deimos on a mask optimized for observing a high redshift supernova identified in a different iscs high redshift cluster ( perlmutter et al .
, in preparation ) which has not yet been spectroscopically confirmed .
four candidate members of iscs j1434.1 + 3328 were targeted , and two more spectroscopic members were identified from additional slitlets targeting 4.5 @xmath71 m selected sources to fill the mask .
all six confirmed members have clearly identified ca h and k absorption lines and d4000 breaks , and none of the galaxies exhibit emission features in the wavelengths covered by the deimos observations ( @xmath151 ) .
iscs j1429.2 + 3357 was observed with deimos , on a slitmask optimized for observing faint , mid - infrared selected agn candidates ( e.g. , * ? ? ?
eight candidate cluster members were included in the slitmask , six of which were confirmed spectroscopically to reside at @xmath152 . only the brightest galaxy , irac j142912.9 + 335808 , shows [ ] emission ; the remaining redshifts were derived on the basis of ca h and k absorption lines and/or d4000 breaks .
in addition , one galaxy targeted spectroscopically as an irac - selected agn candidate , irac j142916.1 + 335537 , is a cluster member , bringing the tally to seven spectroscopically confirmed cluster members .
the spectrum of this source shows strong , narrow ( 400 km s@xmath5 ) emission lines from [ ] @xmath153 , indicating that it is indeed an active galaxy .
iscs j1433.1 + 3334 was observed with deimos on a mask optimized for observing a high redshift supernova identified in iscs j1432.4 + 3332 ( perlmutter et al .
, in preparation ) .
eight candidate members of iscs j1434.1 + 3328 were targeted , of which six were confirmed as cluster members , one was found to be slightly foreground to the cluster , and one was slightly behind the cluster .
all eight candidates show clear d4000 breaks .
many additional cluster members were identified on this mask , either serendipitously or as targeted irac - selected , @xmath0 galaxies , bringing the total number of spectroscopically confirmed cluster members to 20 . as seen in the bottom two panels of figure [ spectra ] ,
the confirmed sources show a range of spectral properties .
while most show ca h and k absorption and d4000 breaks , some also show emission features likely due to either star formation or agn activity .
iscs j1433.2 + 3324 was observed with deimos on the same mask as iscs j1434.1 + 3328 .
five candidate members of iscs j1434.1 + 3328 were targeted , of which two were confirmed as cluster members , one was found to be foreground , and two yielded inconclusive spectra .
four additional cluster members were identified in the same mask , from irac - selected sources .
the two confirmed cluster members which were specifically targeted show strong ca h and k absorption and lack emission lines .
the other four confirmed members all show [ ] emission , with three of the four also showing d4000 breaks .
the spectroscopic observations which confirmed iscs j1432.4 + 3332 at @xmath154 are described in @xcite , but no data on individual sources was presented there . the @xcite result was based on nine spectroscopically confirmed cluster members , two of which were selected not as cluster members , but rather as mid - ir selected agn , and that data is now included in table [ spectable ] of this paper .
one of the candidate members of this cluster hosted a high - redshift supernova ( perlmutter et al . , in preparation ) and was thus targeted for additional deimos and focas slitmask spectroscopy . an example deimos spectrum for this cluster
is shown in the top panel of figure [ spectra ] . in total , there are now 23 spectroscopically confirmed members in the cluster . seven candidate members of this cluster were confirmed spectroscopically during our lris observations in may 2007 .
all seven galaxies show red continuum emission and/or a clear d4000 break .
two of the sources also show [ ] emission .
four of the five spectroscopically confirmed members of this cluster show [ ] emission , a higher proportion than is typical for this program .
this is possibly a selection effect ; sources with line emission are the easiest to spectroscopically confirm .
all five confirmed members show breaks typical of early - type galaxies ( e.g. , d2900 and/or d4000 ) .
this cluster is discussed in @xcite , where eight spectroscopic members were presented .
table [ spectable ] of the current paper does not duplicate those data , and instead lists three additional spectroscopic members that have since been identified .
this cluster has nine spectroscopically confirmed members .
one is an optically - bright agn from the ages survey , while the rest were confirmed spectroscopically by lris .
four of the keck / lris sample show [ ] emission , while the other four show only spectral breaks and absorption lines .
this cluster has eight spectroscopically confirmed members , all from our keck / lris observations in may 2007 .
only one of the sources shows [ ] emission ; the rest of the redshifts are on the basis of continuum breaks at 2640 and 2900 , as well as @xmath155 2800 absorption .
this cluster has five confirmed members , one of which was serendipitously identified in the spectroscopy , all from our keck / lris spectroscopy in june 2005 .
three of the members show [ ] , and the other two show spectral breaks characteristic of early - type galaxy spectra .
this cluster was first published in @xcite , at which time it was the highest redshift galaxy cluster known .
two candidate cluster members hosted supernovae in the _ hst_/acs program of perlmutter et al .
( in preparation ) , and this field thus has been the target of additional spectroscopy from subaru and keck .
five new cluster members have been confirmed , listed in table [ spectable ] .
data on the original five members is not duplicated from @xcite .
of the five new members , one is an [ ] emitter , serendipitously identified in a slitlet targeting another source .
three show only ca h and k absorption lines , with no emission lines identified .
irac j143816.8 + 341440 ( 22.3 in table 3 ) is an agn , showing [ ] , [ ] , and [ ] emission lines . with 20 or more spectroscopic redshifts ,
it is possible to estimate cluster masses via scaling relations using the velocity dispersion . the line of sight velocity dispersion for the 20 spectroscopic member galaxies in cluster 19 ( iscs j1433.1 + 3334 ) at @xmath156 is 760 km s@xmath5 in the rest frame , and for the 23 spectroscopic member galaxies in cluster 17 ( iscs j1432.4 + 3332 ) at @xmath157 it is 734 km s@xmath5 .
for these two clusters , which are among the richest in the sample , the velocity dispersion of @xmath158 km s@xmath5 corresponds to a virial mass of @xmath159 where @xmath160 is the virial radius in mpc ( e.g. equation 4 of * ? ? ? * ) .
the x - ray temperature corresponding to @xmath161 km s@xmath5 is 3.9 kev ( table 4 of * ? ? ?
* ) , which in turn gives a mass of @xmath162 @xcite .
this is consistent with the average halo mass of @xmath163 estimated by @xcite for the sample .
stellar luminosities can also be used to make a rough estimate of cluster masses , scaling to the coma cluster via the red spike model . within the @xmath164 kpc region of coma sampled by @xcite
the integrated @xmath165-band luminosity to @xmath166 is @xmath167 , slightly higher than the values shown in column 11 of table [ z1table ] , but in a significantly smaller effective radius of 0.42 mpc . in their figure 1 ,
geller , diaferio , & kurtz ( 1999 ) show a mass for the coma cluster of @xmath168 within this radius ( for h=0.7 ) . at r=1 mpc
they show about double that mass , and suggest the total mass for coma is about double this again . hence if the profile and @xmath169 for our clusters is similar to coma , allowing for red spike model evolution in @xmath170 , the @xmath171 values in 1 mpc radius provided in table 1 scale to total cluster masses of @xmath172 ( i.e. @xmath173 ) .
clusters 17 and 19 are at the top of this range , again providing mass estimates consistent with those found using the velocity dispersion .
more detailed exploration of this approach will require photometry which takes account of the extent to which a 5 diameter aperture fails to include all of the light from such galaxies , or includes light from multiple galaxies ( as determined from higher spatial resolution imaging ) .
as noted in the introduction , out to @xmath8 substantial evidence exists which is consistent with an extremely simple formation history for cluster galaxies , in which their stars are formed in a short burst at high redshift , and they evolve quiescently thereafter ( ie a `` red spike '' model ) .
the colors of luminous cluster galaxies out to @xmath8 typically fall on a tight sequence which is red relative to field galaxies at similar redshift ( the `` red sequence '' ) , with a mean color which evolves as the red spike model predicts .
the most luminous red sequence galaxies are the reddest , a correlation which is attributed to higher metallicity in higher mass galaxies ( @xcite ; but see also @xcite ) .
this correlation ( i.e the color - magnitude or mass - metallicity relation ) is explained by additional cycles of star formation and enrichment in more massive galaxies , as they are able to retain their gas more effectively against supernovae - driven winds @xcite . in the red spike formation paradigm , this is a natural consequence . in the context of the hierarchical merging galaxy formation models ,
a unique correlation between stellar mass and metallicity is a greater challenge . while the inclusion of feedback ( whether by supernovae or by agn ) in hierarchical models stops the buildup of galaxy mass ( e.g. * ? ? ?
* ; * ? ? ?
* ) , such models have difficulty reproducing the exact color and slope of the color - magnitude relation in clusters over the full redshift range for which it has been measured .
recent work shows promise , however , as feedback behavior and other `` gastrophysical '' effects are taken more into account ( e.g. * ? ? ?
if the onset of the star formation spike " is simultaneous for all cluster galaxies , and if the color - magnitude relation is caused by more protracted spikes in more massive galaxies , then as one approaches the star - forming epoch , massive galaxies might be expected to eventually become bluer , reversing the slope of the color - magnitude relation . in fact no measurable change is seen in color - magnitude slope out to @xmath8 ( e.g * ? ? ?
* ; * ? ? ?
* ) , which suggests either formation redshifts well before @xmath174 , or that the spikes begin earlier in more massive galaxies , perhaps ending rather than beginning simultaneously . the small and unchanging scatter of the red sequence in clusters out to @xmath175 ( e.g. *
* ; * ? ? ?
* ; * ? ? ?
* ) also argues for synchronized or very early spikes and against a primarily age - based origin of the color - magnitude relation , since the scatter would increase by a much larger amount , and faster , than has been observed .
testing to what extent red spike models remain consistent with cluster galaxy data at @xmath0 is one of the primary motivations for the present study .
we begin by constructing the color - magnitude relations for the 12 spectroscopically confirmed clusters in table [ z1table ] .
figures [ cmds1 ] and [ cmds2 ] present color magnitude diagrams for the clusters listed in table [ z1table ] . the @xmath7 $ ] color was selected because these filters bracket the 4000 break most tightly at @xmath0 .
the symbol area is proportional to the integral of the object s redshift probability distribution over the range @xmath176 , which is the rms dispersion in individual photometric redshifts @xcite .
circled symbols are spectroscopically confirmed members , while crosses indicate objects known _ not _ to be members on the basis of spectroscopy .
figures [ cmds1 ] and [ cmds2 ] include objects within 1 mpc of the cluster center , with @xmath177 detections in both [ 3.6 ] and [ 4.5 ] , and with @xmath178 detections in @xmath57 ; or with spectroscopic redshifts . as noted in [ sec : phot - limits ] , the @xmath92 limit in [ 3.6 ] is 18.6 mag , so the limiting factor in object selection for figures [ cmds1 ] and [ cmds2 ] is from the @xmath92 limit of 17.8 mag in [ 4.5 ] . we have used the 0.1 gyr burst , @xmath9 red spike model shown in figure [ m_vs_z ] to calculate the corresponding [ 3.6 ] mag as a function of redshift , and this is shown in figures [ cmds1 ] and [ cmds2 ] by the vertical dotted lines .
the @xmath57 limit is shown by the diagonal dotted lines in figures [ cmds1 ] and [ cmds2 ] .
the vertical dashed line plots the expected l * magnitude at [ 3.6 ] for the red spike model .
the sloped dashed line shows the observed @xmath179 color - magnitude slope of 0.22 for the coma cluster from the data of @xcite , normalized at the red spike model [ 3.6 ] magnitude and @xmath7 $ ] color for an l * galaxy . in this redshift range
observed @xmath7 $ ] is close to rest @xmath179 .
note that while the normalization evolves according to the red spike model , a constant , _ unevolving _ slope value is used .
the brightest galaxy likely to be a member is typically 1 to 2 magnitudes brighter than the expected l * magnitude for the red spike model ( see table [ z1table ] ) .
it is noteworthy that the brightest galaxy in the coma cluster is @xmath180 times brighter than l * in the @xmath41 band @xcite , suggesting less than a factor of two growth in stellar mass in such galaxies since @xmath181 .
_ hst _ imaging should be used to assess this more carefully . because these clusters were _
not _ selected on the basis of containing a red sequence ( [ sec : detn ] ) , the fact that the highest probability cluster members tend to track the passively evolving coma cluster sequence shown by the dashed line is significant .
evidently the red sequence persists in dense environments to @xmath182 , even when not used as a selection criterion .
in addition there is no indication that the slope of the color - magnitude relation has changed sign , and in fact a non - evolving coma cluster slope appears consistent with the data .
comparing the data for e.g. cluster 14 ( iscs j1426.5 + 3339 at @xmath183 ) vs. 29 ( iscs j1432.6 + 3436 at @xmath184 ) suggests real differences in the scatter of the color - magnitude relation do exist .
the presence of luminous galaxies substantially bluer than the red sequence in several clusters , a number of them spectroscopically confirmed , is also noteworthy .
these may represent a population of massive star - forming galaxies which fade onto the red sequence in rich clusters , but is not found in field surveys ( e.g. * ? ? ?
it should be noted that the very low scatters reported for @xmath185 clusters in e.g. @xcite and @xcite are calculated for galaxies known to have early - type morphologies , using much deeper photometry than the survey / discovery data used here .
further investigation of the color - magnitude relations is deferred to a future paper which will use deeper _ spitzer _ and _ hst _ imaging presently being obtained .
although the survey data used to identify the clusters does not enable accurate photometry of individual galaxies in clusters , we can calculate mean properties for galaxies in each cluster with some confidence . in the next section
we briefly explore the color evolution and color - magnitude relation of the entire cluster sample .
figure [ im3p6vsz ] plots the average @xmath7 $ ] color for galaxies within 1 mpc of the cluster centers , and whose integrated redshift probability distribution in the range @xmath176 exceeds 0.3 . the mean
values are calculated after iteratively clipping @xmath186 outliers . as implied by the color - magnitude diagrams for the spectroscopically confirmed clusters ( [ sec : cmds ] ) ,
the red spike model ( 0.1 gyr burst at @xmath9 ) provides a remarkably good fit to the empirical data , validating the survey design assumptions illustrated in figure [ m_vs_z ] .
much of the increase in @xmath7 $ ] with redshift is due to the change in the rest frame wavelengths observed , and in figure [ im3p6relnevsz ] we plot the average color relative to a no - evolution model ( i.e. the k - corrected color ) .
the no - evolution model is simply the red spike model at z=0 , i.e. at an age of 11.3 gyr for the stellar population .
clearly the evolving red spike model is a better fit to the data than is the k - correction alone .
note that the photometric redshifts ( [ sec : photz ] ) are calculated using _
non_-evolving templates , so this result is not a foregone conclusion , particularly at @xmath0 .
the mean @xmath7 $ ] colors in figure [ im3p6vsz ] include galaxies down to the @xmath187jy survey limit at @xmath1 m , and therefore less luminous galaxies contribute to the mean at lower redshifts .
the color - magnitude relation shows that less luminous galaxies are bluer , and massive galaxies are quite quiescent today , so a constant flux limit will lead to a systematic bias towards more luminous , massive galaxies and redder mean color with increasing redshift . on the other hand
the excellent fit of the red spike model implies that a constant _ luminosity _
limit for the sample would select galaxies with smaller stellar masses at high redshift , since a constant stellar mass , passively evolving , will be more luminous as one approaches the formation redshift . to avoid this bias , in figure [ im3p6relnevsz ]
we also show ( with red squares ) the mean colors for galaxies brighter than the red spike passively - evolving l * in [ 3.6 ] .
it is evident that the mean color of the more luminous galaxies is systematically redder , and hence that the color - magnitude relation continues to hold out to @xmath181 .
the color offset is 0.1 magnitudes , independent of redshift .
the persistence of the essentially unchanged color - magnitude relation slope , with a passively evolving intercept , out to lookback times within 4 gyr of the big bang is a phenomenon that models of cluster galaxy formation must account for .
the implication is that the correlation between high stellar population metallicity and stellar mass is already in place at @xmath181 , and that the star formation era remains well in the past for these galaxies . in figure [ im3p6relpevsz ]
the mean @xmath7 $ ] colors are plotted with the predicted @xmath7 $ ] color for the red spike model subtracted .
the mean colors are for galaxies more luminous in [ 3.6 ] than the red spike l*. note the red spike model is a good match to @xmath54 galaxy colors in figures [ im3p6vsz ] and [ im3p6relnevsz ] , so the offset in color out to @xmath8 is attributable to the color - magnitude relation .
other effects may be in play . in table 1
the @xmath117 values tend to be lower than the @xmath188 values , which may lead to colors appearing @xmath189 mag redder relative to the red spike model than if @xmath188 were available for all @xmath0 clusters .
the average colors of cluster members may still be redder than shown , however , as no attempt has been made to correct the average cluster colors for field contamination , and field galaxies tend to be bluer than cluster galaxies .
blending issues could lead to colors which are systematically biased .
given these uncertainties , the excellent agreement with the red spike model is the more remarkable .
nevertheless , at @xmath0 the trend is towards colors which are increasingly redder than the red spike model , for both the full sample and the spectroscopically confirmed subset , though there is clearly a range .
hence even higher formation redshifts ( as high as @xmath190 ) are favored for most @xmath0 clusters , within the context of red spike models .
figure [ im3p6vsz ] shows that that there are relatively few cluster candidates in the irac sample with @xmath192 .
why is this ?
the red spike model fits , circumstantial evidence from bcg luminosity , and the persistence of the color - magnitude relation suggest that the stellar populations were formed and assembled
well before @xmath181 . from figure [ m_vs_z ] , such clusters should be detectable by the irac shallow survey .
a potential selection effect against @xmath115 galaxies is that photometric redshifts in this range have increasingly broad redshift probability distributions , due in part to insufficient depth in photometry at shorter wavelengths .
simulations indicate a factor of two reduction in irac photometric error can help compensate , leading to a similar reduction in photometric redshift error in this redshift range . with the _ spitzer _ deep wide - field survey legacy program now underway ( pi d. stern ) , such data will be available in the near future .
the tighter redshift probability distributions would improve the contrast of high redshift clusters over the field , allowing them to meet our detection threshold .
but the lack of massive @xmath115 cluster detections may simply reflect a real decline in the space density of such objects .
models for the hierarchical growth of structure predict only about 1 cluster with @xmath193 above @xmath194 for every 5 clusters with @xmath0 .
full exploration of this possibility will require more careful simulation and assessment of our observational selection effects .
we have identified 335 galaxy cluster and group candidates from a @xmath53 m selected sample of galaxies in the irac shallow survey .
candidates were identified by searching for overdensities in photometric redshift slices , and 106 clusters are at @xmath0 .
roughly 10% of these candidates may be expected to arise by chance or from projection effects . to date , 12 clusters have been spectroscopically confirmed at @xmath0 , as have 61 of the 73 clusters observed with ages at @xmath125 .
for the two @xmath0 clusters with 20 or more spectroscopic members , total cluster masses of several @xmath195 are indicated , and the total mass estimated from the stellar luminosity yields comparable values .
color - magnitude diagrams in @xmath7 $ ] vs. [ 3.6 ] for the @xmath0 spectroscopically confirmed clusters reveal that a red sequence is generally present , even though clusters were not selected for this .
the brightest probable member galaxy ( at the spatial resolution of irac ) in the spectroscopically confirmed @xmath185 clusters remains 1 2 mag brighter than the passively - evolving @xmath54 luminosity .
for the full cluster sample , the mean color of brighter galaxies within each cluster is systematically redder than the mean color of all probable cluster member galaxies , implying that the mass - metallicity relation is already in place at @xmath181 .
the mean @xmath7 $ ] color of probable cluster members is well fit by a simple model in which stars form in a 0.1 gyr burst beginning at @xmath9 . at @xmath0
there is a tendency for mean cluster colors to favor formation redshifts @xmath10 , although a few are consistent with @xmath196 .
this adds to the large body of evidence that galaxies in clusters were established at extremely early times .
we thank mark dickinson , emily macdonald , and hyron spinrad for generously making time available on their scheduled nights for the deimos observations reported here .
naoki yasuda , naohiro takanashi , yutaka ihara , kohki konishi , and hiroyuki utsunomiya assisted with observations at the subaru telescope .
roberto de propris provided the integrated luminosity for coma cluster galaxies .
thoughtful comments from the anonymous referee improved the presentation of this work .
the irac shallow survey was executed using guaranteed observing time contributed by g. fazio , g. and m. rieke , m. werner , and e. wright .
this work is based in part on observations made with the _
spitzer space telescope _ , which is operated by the jet propulsion laboratory , california institute of technology under a contract with nasa .
this work made use of images and data products provided by the noao deep wide - field survey , which is supported by the national optical astronomy observatory ( noao ) .
noao is operated by aura , inc .
, under a cooperative agreement with the national science foundation .
some of the data presented herein were obtained at the w.m .
keck observatory , which is operated as a scientific partnership among the california institute of technology , the university of california and the national aeronautics and space administration .
the observatory was made possible by the generous financial support of the w.m .
keck foundation .
some of the data presented were collected at the subaru telescope , which is operated by the national astronomical observatory of japan .
the work of sas was performed under the auspices of the u.s .
department of energy , national nuclear security administration , by the university of california , lawrence livermore national laboratory , under contract no .
the work of kd and jm was partially supported by the director , office of science , department of energy , under grant de - ac02 - 05ch11231 .
lccc + 152.1 & 15.45 & 1.01 & 1.057 + 152.2 & 15.99 & 0.91 & 1.055 + 152.3 & 17.00 & 0.96 & 1.054 + 152.4 & 17.47 & 1.04 & 1.065 + 152.5 & 16.40 & 0.99 & 1.055 + 152.6 & 16.76 & 0.96 & 1.055 + + 51.1 & 15.10 & 0.98 & 1.059 + 51.2 & 17.06 & 0.95 & 1.056 + 51.3 & 17.83 & 1.04 & 1.055 + 51.4 & 17.16 & 1.00 & 1.060 + 51.5 & 17.01 & 1.03 & 1.059 + 51.6 & 17.19 & 1.03 & 1.054 + 51.7 & 16.08 & 1.01 & 1.060 + + 19.1 & 17.49 & 0.99 & 1.075 + 19.2 & 16.67 & 0.99 & 1.075 + 19.3 & 17.04 & 1.06 & 1.076 + 19.4 & 16.00 & 1.06 & 1.066 + 19.5 & 16.89 & 1.09 & 1.067 + 19.6 & 16.47 & 0.94 & 1.065 + 19.7 & 17.70 & 1.00 & 1.063 + 19.8 & 17.15 & 1.10 & 1.079 + 19.9 & 16.75 & 0.98 & 1.074 + 19.10 & 16.30 & 1.07 & 1.066 + 19.11 & 16.21 & 1.03 & 1.066 + 19.12 & 16.66 & 1.03 & 1.075 + 19.13 & 15.44 & 1.06 & 1.064 + 19.14 & 17.03 & 1.28 & 1.076 + 19.15 & 17.51 & 1.00 & 1.065 + 19.16 & 17.01 & 1.27 & 1.069 + 19.17 & 17.36 & 1.04 & 1.066 + 19.18 & 16.27 & 0.86 & 1.063 + 19.19 & 16.68 & 2.45 & 1.074 + 19.20 & - & - & 1.067 + + 123.1 & 15.32 & 1.07 & 1.094 + 123.2 & 16.45 & 1.09 & 1.093 + 123.3 & 16.68 & 1.10 & 1.094 + 123.4 & 16.99 & 1.11 & 1.107 + 123.5 & 16.64 & 0.93 & 1.094 + 123.6 & 17.35 & 0.99 & 1.091 + + 17.1 & 16.94 & 2.74 & 1.1120 + 17.2 & 17.35 & 1.03 & 1.1108 + 17.3 & 16.48 & 1.07 & 1.111 + 17.4 & 15.97 & 1.05 & 1.115 + 17.5 & 16.31 & 1.04 & 1.111 + 17.6 & 15.85 & 1.09 & 1.110 + 17.7 & 16.04 & 1.35 & 1.121 + 17.8 & 17.57 & 1.20 & 1.116 + 17.9 & 16.87 & 1.00 & 1.11 + 17.10 & 17.31 & 1.04 & 1.1086 + 17.11 & 17.01 & 1.14 & 1.098 + 17.12 & 17.43 & 1.12 & 1.105 + 17.13 & 17.51 & 0.98 & 1.109 + 17.14 & 17.32 & 0.91 & 1.112 + 17.15 & 16.59 & 1.07 & 1.104 + 17.16 & 16.92 & 1.19 & 1.119
+ 17.17 & 17.42 & 1.09 & 1.115 + 17.18 & 17.33 & 1.11 & 1.115 + 17.19 & 17.29 & 1.06 & 1.118 + 17.20 & 16.90 & 1.18 & 1.107 + 17.21 & 16.38 & 1.26 & 1.115 + 17.22 & - & - & 1.114 + 17.23 & - & - & 1.110 + + 34.1 & 16.35 & 1.03 & 1.1439 + 34.2 & 15.75 & 1.04 & 1.1301 + 34.3 & 16.01 & 1.11 & 1.1271 + 34.4 & 15.69 & 1.00 & 1.1328 + 34.5 & 16.12 & 1.18 & 1.134 + 34.6 & 16.90 & 0.91 & 1.13 + 34.7 & 17.24 & 0.89 & 1.144 + + 14.1 & 16.87 & 1.10 & 1.16 + 14.2 & 15.96 & 1.05 & 1.157 + 14.3 & 16.44 & 1.19 & 1.1631 + 14.4 & 16.64 & 1.17 & 1.1634 + 14.5 & 16.88 & 1.04 & 1.1637 + + 342.1 & 16.76 & 1.21 & 1.240 + 342.2 & 17.71 & 1.21 & 1.251 + 342.3 & - & - & 1.256 + + 30.1 & 16.65 & 1.18 & 1.245 + 30.2 & 16.39 & 1.13 & 1.26 + 30.3 & 17.24 & 1.15 & 1.2576 + 30.4 & 17.37 & 1.07 & 1.263 + 30.5 & 17.41 & 1.29 & 1.2611 + 30.6 & 16.86 & 1.08 & 1.2583 + 30.7 & 16.97 & 1.31 & 1.2582 + 30.8 & 15.19 & 0.97 & 1.2632 + 30.9 & - & - & 1.2546 + + 29.1 & 17.63 & 1.26 & 1.3559 + 29.2 & 17.61 & 1.25 & 1.35 + 29.3 & 16.39 & 1.32 & 1.35 + 29.4 & 16.06 & 1.21 & 1.3320 + 29.5 & 17.55 & 1.40 & 1.347 + 29.6 & 16.77 & 1.29 & 1.347 + 29.7 & 16.86 & 1.30 & 1.34 + 29.8 & 16.14 & 1.26 & 1.353 + + 25.1 & 17.52 & 3.33 & 1.37 + 25.2 & 16.98 & 1.53 & 1.372 + 25.3 & 16.28 & 1.34 & 1.37 + 25.4 & 15.68 & 1.33 & 1.374 + 25.5 & - & - & 1.380 + + 22.1 & 15.86 & 1.25 & 1.411 + 22.2 & 15.79 & 1.38 & 1.418 + 22.3 & 16.91 & 1.38 & 1.412 + 22.4 & 16.76 & 1.35 & 1.414 + 22.5 & - & - & 1.412 + | we have identified 335 galaxy cluster and group candidates , 106 of which are at @xmath0 , using a @xmath1 m selected sample of objects from a 7.25 deg@xmath2 region in the _ spitzer _ infrared array camera ( irac ) shallow survey .
clusters were identified as 3-dimensional overdensities using a wavelet algorithm , based on photometric redshift probability distributions derived from irac and noao deep wide - field survey data .
we estimate only @xmath3 of the detections are spurious . to date 12 of the @xmath0 candidates
have been confirmed spectroscopically , at redshifts from 1.06 to 1.41 .
velocity dispersions of @xmath4 km s@xmath5 for two of these argue for total cluster masses well above @xmath6 , as does the mass estimated from the rest frame near infrared stellar luminosity .
although not selected to contain a red sequence , some evidence for red sequences is present in the spectroscopically confirmed clusters , and brighter galaxies are systematically redder than the mean galaxy color in clusters at all redshifts .
the mean @xmath7 $ ] color for cluster galaxies up to @xmath8 is well matched by a passively evolving model in which stars are formed in a 0.1 gyr burst starting at redshift @xmath9 . at @xmath0 , a wider range of formation histories is needed , but higher formation redshifts ( i.e. @xmath10 ) are favored for most clusters . |
we are grateful to the u.s .
department of energy for financial support . | this paper demonstrates that complex @xmath0-symmetric periodic potentials possess real band spectra
. however , there are significant qualitative differences in the band structure for these potentials when compared with conventional real periodic potentials . for example , while the potentials @xmath1 @xmath2 have infinitely many gaps , at the band edges there are periodic wave functions but no antiperiodic wave functions . numerical analysis and higher - order wkb techniques
are used to establish these results .
.5 cm for a quantum mechanical model having a periodic potential the schrdinger equation is @xmath3 where the potential @xmath4 is periodic with period @xmath5 : @xmath6 in conventional treatments of eq .
( [ e1 ] ) @xcite the periodic potential @xmath4 is assumed to be real . imposing the condition that the wave function @xmath7 be bounded leads to a real spectrum consisting of continuous bands separated by gaps .
there is an infinite number of bands and gaps , except for the special family of so - called _ finite - gap _ potentials such as the lam potentials @xcite . in this note
we extend the conventional analysis to include the case of complex periodic potentials .
we find that complex periodic potentials having @xmath0 symmetry exhibit _ real _ band spectra , despite the non - hermitian character of the schrdinger equation ( [ e1 ] ) .
( here , @xmath8 represents parity reflection and @xmath9 represents time reversal . ) potentials having this symmetry satisfy @xmath10^*=v(x ) .
\label{e3}\end{aligned}\ ] ] examples of such potentials are @xmath11 , @xmath12 , and @xmath13 .
in addition to the property that these potentials have real spectra , the band structure displays several novel features that are strikingly different from the case of real periodic potentials .
the work reported here was motivated by recent investigations of non - hermitian @xmath0-symmetric hamiltonian models having real discrete spectra .
one such class of models is defined by the hamiltonian @xcite @xmath14 despite the lack of conventional hermiticity , the spectrum of this hamiltonian is real , positive , and discrete ; each of the energy levels increases as a function of increasing @xmath15 .
it has been observed that the reality of the spectrum is a consequence of @xmath0 symmetry , which is a weaker condition than hermiticity .
this observation has also been used to construct new classes of quasi - exactly solvable quantum theories @xcite and to study new kinds of symmetry breaking in quantum field theory @xcite .
there have been many other instances of non - hermitian @xmath0-invariant hamiltonians in physics .
energies of solitons on a _ complex _ toda lattice have been found to be real @xcite .
hamiltonians rendered non - hermitian by an imaginary external field have been used to study population biology @xcite and to study delocalization transitions such as vortex flux - line depinning in type - ii superconductors @xcite . in these last two cases ,
initially real eigenvalues bifurcate into the complex plane due to the increasing external field , indicating the growth of populations or the unbinding of vortices .
we begin by summarizing the standard floquet analysis of the schrdinger equation ( [ e1 ] ) for the case where @xmath4 is real and periodic @xcite .
we define a _ fundamental pair _ of linearly independent solutions @xmath16 and @xmath17 satisfying the initial conditions @xmath18 any solution @xmath7 to eq .
( [ e1 ] ) is a linear combination of @xmath16 and @xmath17 .
it is then a straightforward algebraic exercise to show that @xmath7 is bounded provided that the _ discriminant _ @xmath19 , which is defined by @xmath20 satisfies the constraint @xmath21 to illustrate the features of the discriminant we consider a typical periodic potential , @xmath22 , for which the period @xmath23 . in fig .
[ f1 ] we plot @xmath19 as a function of @xmath24 .
note that @xmath19 is oscillatory and is well approximated by the function @xmath25 for large @xmath24 .
the crucial feature of @xmath19 , which can not be seen from this plot , is that its graph crosses the lines @xmath26 an infinite number of times ; each of the maxima of @xmath19 lies above @xmath27 and each of the minima lies below @xmath28 .
the regions of energy for which @xmath29 are called _ bands _ and the regions of energy for which @xmath30 are called _ gaps_. the gap size decreases exponentially as a function of @xmath24 .
the band edges at which @xmath31 correspond to periodic solutions to eq .
( [ e1 ] ) , @xmath32 , and the band edges at which @xmath33 correspond to antiperiodic solutions @xmath34 . now consider the calculation of the discriminant for the case of a _ complex _ periodic potential @xmath4 .
in general , a complex periodic potential will have no bounded solutions because the discriminant is typically complex . however , for complex @xmath0-symmetric periodic potentials , one can easily show that the discriminant @xmath19 is real when @xmath24 is real .
the @xmath0 symmetry is crucial here ; for a potential that is not @xmath0 symmetric [ one that does not satisfy eq .
( [ e3 ] ) ] , the discriminant is complex for all values of @xmath24 .
having established that complex @xmath0-symmetric periodic potentials have real discriminants , we can then apply the criterion in eq .
( [ e7 ] ) to locate the real energy bands within which the corresponding wave function @xmath7 is a bounded function .
we have computed the discriminants for the class of complex @xmath35-symmetric periodic potentials @xmath36 in figs .
[ f2]-[f7 ] we plot the discriminants for the cases @xmath37 . while these plots superficially resemble fig .
[ f1 ] for large @xmath24 , they exhibit new and intriguing features that are significantly different from the case of a real periodic potential .
the most obvious new feature is the appearance for @xmath38 of a local minimum of @xmath19 between @xmath28 and @xmath27 .
such a dip is rigorously forbidden in the case of real periodic potentials @xcite . the most dramatic differences between the discriminants for the complex @xmath35-symmetric periodic potentials in eq .
( [ e8 ] ) and real periodic potentials can not be easily seen in the figures .
we have performed a careful numerical study of the local minima and maxima of @xmath19 .
our study reveals that _ none of the local minima lies below _
@xmath28 .
this shows that there are _ no antiperiodic solutions _
@xmath7 to the schrdinger equation ( [ e1 ] ) .
nevertheless , all of the local maxima of @xmath19 lie above @xmath27 .
hence , there are an infinite number of band gaps in the spectrum and the band - edge wave functions are periodic . @xmath39 @xmath39 to perform this numerical analysis it is necessary to locate the positions of the local minima and maxima of @xmath19 to extremely high accuracy
. this can be done using wkb methods @xcite .
we take the energy @xmath24 to be large @xmath40 and define a small parameter @xmath15 by @xmath41 then we make an exponential _ ansatz _ for the wave function @xmath7 : @xmath42 . \label{e10}\end{aligned}\ ] ] substituting @xmath7 in eq .
( [ e10 ] ) into the schrdinger equation ( [ e1 ] ) gives a recursion relation for the functions @xmath43 : @xmath44 ^ 2&=&0,\nonumber\\ iq_0''(x)-2q_0'(x)q_1'(x)&=&0,\nonumber\\ iq_1''(x)-2q_0'(x)q_2'(x)-[q_1'(x)]^2-v(x)&=&0,\nonumber\\ iq_{n-1}''(x)-\sum_{j=0}^nq_j'(x)q_{n - j}'(x)&=&0\quad(n\geq3 ) .
\label{e11}\end{aligned}\ ] ] the solution to these equations is @xmath45,\nonumber\\
q_4(x)&=&\pm{1\over8}\left(v'(x)-v'(0)-\int_0^xdt\,v^2(t)\right),\nonumber\\ q_5(x)&=&{i\over16}\left[v''(x)-v''(0)-2v^2(x)+2v^2(0)\right],\nonumber\\ q_6(x)&=&\mp{1\over32}\left(v'''(x)-v'''(0)-5v(x)v'(x)+5v(0)v'(0 ) + \int_0^xdt\,[2v^3(t)-v(t)v''(t)]\right ) , \label{e12}\end{aligned}\ ] ] and so on .
note that @xmath43 is normalized so that @xmath46 .
in general , the formula for @xmath47 is @xmath48\quad(n\geq3 ) .
\label{e13}\end{aligned}\ ] ] in order to obtain a wkb formula for the discriminant @xmath19 in eq .
( [ e6 ] ) we need to evaluate @xmath43 at @xmath49 .
the periodicity of the potential @xmath4 simplifies the results considerably ; when @xmath50 is odd , @xmath51 and when @xmath50 is even , only the integrals in eq .
( [ e12 ] ) remain : @xmath52 , \label{e14}\end{aligned}\ ] ] and so on .
the wkb formula for the discriminant is particularly simple when the potential @xmath4 is @xmath0 symmetric : @xmath53 . \label{e15}\end{aligned}\ ] ] one obtains the same formula for potentials that are real and symmetric under parity @xmath8 . for the complex @xmath0-symmetric potentials @xmath4 in ( [ e8 ] ) the wkb formula for the discriminant in eq .
( [ e15 ] ) is @xmath54 .
\label{e16}\end{aligned}\ ] ] a similar wkb formula exists for the real odd - parity potentials @xmath55 : @xmath56 . \label{e17}\end{aligned}\ ] ] we can illustrate the extreme accuracy of these wkb approximations by comparing them with numerical computations of the discriminant .
for example , for @xmath57 at @xmath58 ( which corresponds to @xmath59 ) we find numerically that @xmath60 .
the first three orders of the wkb approximation taken from eq .
( [ e16 ] ) give @xmath61 , @xmath62 , and @xmath63 .
similarly , for @xmath64 at @xmath58 we find numerically that @xmath65 .
the first three orders of the wkb approximation taken from eq .
( [ e17 ] ) give the same values : @xmath66 , @xmath62 , @xmath63 .
despite this impressive precision , the wkb formulas ( [ e16 ] ) and ( [ e17 ] ) can not be used directly to answer the crucial question of whether there are band gaps because these approximations to the discriminant @xmath19 never cross the values @xmath26 .
the reason for this inadequacy of the wkb approximation is that the differences @xmath67 and @xmath68 are exponentially small when @xmath69 .
therefore , these differences are subdominant with respect to the wkb asymptotic series and are not accessible to any order in powers of @xmath15 . indeed
, the wkb series can only provide information about quantities with an _ algebraically _ small error , and not an exponentially small error .
we emphasize that the wkb approximation has this shortcoming only at the maxima and minima of the approximation . at other points
any exponential discrepancy is completely negligible compared with algebraic errors .
the wkb series is still an extremely useful ingredient in the numerical search for zeros of @xmath70 .
( these zeros are the dividing points between bands and gaps . )
our procedure is first to find the energies at which there are maxima and minima of the wkb approximation to the discriminant and then to evaluate , with high numerical precision , the actual value of the discriminant in a tiny neighborhood of each of these points . by doing this
we can determine whether or not the discriminant @xmath19 crosses the lines @xmath26 .
for the real periodic potentials @xmath71 our procedure confirms the rigorous theoretical result that every maximum of the discriminant lies above @xmath27 and every minimum lies below @xmath28 . consider , for example , the potential @xmath22 . from fig .
1 , it is clear that the first maximum lies above @xmath27 .
the first minimum occurs at @xmath72 , where the discriminant has the value @xmath73 .
the second maximum occurs at @xmath74 , where the discriminant is @xmath75 .
similar behavior is found for the other potentials in the class @xmath71 . in stark contrast , for the potentials @xmath76 , while the maxima of the discriminant lie above @xmath77 , the minima of the discriminant lie above @xmath28 .
thus , for these potentials there are no antiperiodic wave functions . as an example
, lengthy and delicate numerical analysis verifies that for the potential @xmath78 the first three maxima of the discriminant @xmath19 are located at @xmath79 , @xmath80 , and @xmath81 .
the value of the discriminant @xmath19 at these energies is @xmath75 , @xmath82 , and @xmath83 .
the first two minima of the discriminant are located at @xmath84 and @xmath85 and at these energies @xmath19 has the values @xmath86 and @xmath87 .
similar behavior is found for the other potentials in the class ( [ e8 ] ) .
we conclude by pointing out that from the expressions for @xmath88 in eq .
( [ e14 ] ) the wkb series truncates if the potential is a polynomial in @xmath13 .
for example , for the complex @xmath0-symmetric periodic potential @xmath89 the wkb series in eq .
( [ e15 ] ) truncates after the first term because @xmath90 vanishes for @xmath91 . for this case
the wkb approximation is exact and the discriminant is given by @xmath92 one can verify this result directly by solving the schrdinger equation ( [ e1 ] ) for this potential exactly ; the solution is a bessel function : @xmath93 . |
in molecular clouds , clumps of gas can become gravitationally bound and contract . ultimately , the compressing gas heats up sufficiently to drive the fusion of hydrogen into helium and the birth of a star . for most of its life , hydrogen fusion powers the star ( the so - called main sequence evolution ) .
ultimately , the hydrogen is depleted in the core and the star expands into a giant phase as its core contracts . for massive stars , this phase represents just the first in a series of burning phases where the ashes of the first phase ignite as the core contracts : core contraction continues until helium fusion occurs , a second core contraction until the products of helium fusion ( carbon , oxygen ) ignite , and so on until an iron core is built in the center of the star .
silicon shell burning above the iron core continues to increase its mass until the thermal and electron degeneracy pressure in the core is no longer able to support its mass and the core collapses .
the collapse of these cores produce most of the stellar - massed neutron stars ( nss ) and black holes ( bhs ) in the universe .
the energy released in these core collapses produce supernovae ( type ib / c and ii ) and long - duration gamma - ray bursts ( grbs ) .
the nature and mass of the remnants is directly connected to the nature of the explosion . in this paper , we review the current models for the explosions from stellar implosion and the resultant role they play on the compact remnants produced in the implosion . section [ sec : explosion ] reviews the current explosion mechanisms and section [ sec : bhformation ] reviews how these explosions dictate the remnant mass distribution .
these sections cover theoretical expectations .
we conclude with a discussion of the current observations including the expectations from gravitational wave detections .
as the iron core collapses , the subsequent higher densities and temperatures drive further dissociation of the iron in the core ( removing thermal support ) and increased electron capture ( removing electron degeneracy support ) causing the collapse to accelerate
. ultimately , the core collapses at over 1/10th the speed of light .
this nearly free - fall collapse continues until the core reaches nuclear densities , where nuclear forces and neutron degeneracy halt the collapse , producing a proto - neutron star ( pns ) , the high entropy cores of very massive stars , the collaps halts before reaching nuclear densities and the core can collapse to a bh prior to reaching nuclear densities . ] . at this time , a roughly 0.8 - 1.0m@xmath0 iron core has collapsed from @xmath1 km down to @xmath2 km .
the sudden halt of the collapse causes the infalling material to `` bounce '' . in the collapse , roughly @xmath3erg is the gravitational constant , @xmath4 is the mass of the core ( 1 - 1.4m@xmath0 ) and @xmath5 is the radius of the collapsed core ( typically this radius is initially 50 km but as the pns cools , it contracts to 10 km ) . ] of gravitational potential energy is released ( far above what is needed to power typical supernovae ) .
unfortunately , converting just 1% of this energy into explosion energy to make a @xmath6erg supernovae has proven extremely difficult .
the first simulations argued that the `` bounce '' would convert the potential energy released into outward - moving kinetic energy , driving an explosion @xcite .
however , much of the energy in this bounce is stored in thermal / neutrino energy and when neutrino transport is included , the bounce explosion stalls as soon as the neutrinos are no longer trapped in the shock ( roughly at 100 - 200 km ) . over the past 40 - 50 years , astrophysicists have focused on studying how to revive this shock .
the rest of the star falls onto the stalled shock and the ram pressure of this infalling material must be overcome to drive a successful explosion .
both a region within the pns and the region between the pns and the stalled shock are susceptible to a number of convective instabilities .
these instabilities can increase the efficency at which the potential energy is converted into explosion energy to revive the shock and blow off the infalling star .
much of the current work is focusing on the interplay of the enhancement in energy conversion from convection coupled to the increasingly complex physics needed to model core collapse ( behavior of matter at extreme densities , neutrino emission and cross - section , neutrino transport , and magnetic fields ) .
most of the current work is focused on the convection between the pns and the stalled shock @xcite , where turbulence is able to extract energy released near the surface of the ns and convert it to kinetic energy driving out the shock ( see figure [ fig : snconvdiag ] ) .
the energy in the convective region must overcome the ram pressure of the infalling star .
this ram pressure ( @xmath7 ) of the infalling material rapidly decreases with radius ( @xmath8@xcite ) , so if the stalled shock can be forced outward , it will quickly accelerate , launching an explosion .
our current focus is on understanding the nature of the turbulence ( e.g. whether it is driven by rayleigh taylor or standing accretion shock instabilities ) , the source of the energy ( hot proto - neutron star , pns oscillations , pressure waves from material accreting onto the pns ) and what critical physics is missing in the current models ( e.g. neutrino interactions and transport , equation of state , magetic fields ) .
all these topics remain areas of intense debate .
fueled by the fact that grbs and hypernovae are almost certainly driven by a different mechanism , many astronomers are studying mechanisms beyond this convection - enhanced engine . before we discuss these alternate mechanisms , let s study in more detail some of the scenarios within the convection - enhanced engine paradigm using figure [ fig : snconvdiag ] as a guide .
after the collapse and bounce , unless there are significant pathologies in the equation of state , the entropy profile from the pns out to the shock will decrease ( @xmath9@xcite and such a profile is unstable to rayleigh - taylor instabilities@xcite .
the growth time of such instabilities is roughly a few milliseconds@xcite and we expect rayleigh - taylor to dominate at early times .
much of the potential energy released in the collapse of the core is converted into heat in the pns and this heat leaks of of this star via neutrinos .
these neutrinos heat material just above the pns , further increasing the entropy gradient to drive further rayleigh - taylor convection .
the neutrino - deposited energy is converted into kinetic energy as the hot bubbles rise from the pns base , pushing against the infalling star .
this convective engine is depicted in a slice of a 3-dimensional calculation in figure [ fig : snsim ] .
as the collapsing star piles up on the stalled shock , it flows toward low points , piling up until a downflow is produced .
the low entropy material flows down onto the pns , producing shocks and enhancing the neutrino emission ( contributing additional sources of energy ) .
if the energy is sufficient to drive the shocked region outward , the ram pressure will quickly decrease , allowing a strong supernova explosion .
as the explosion becomes strong , the convective engine turns off . in this way
, the amount of energy in the explosion can be estimated by setting it equal to the energy stored in the convective region when the ram - pressure lid is blown off , allowing us to calculate fallback and final remnant masses ( sec .
[ sec : bhformation ] ) .
rayleigh - taylor convection should occur in all simulations of stellar collapse .
indeed , calculations that do nt exhibit rayleigh - taylor convection typically either have some pathology in their equation of state preventing a large initial entropy gradient , their resolution is too low and numerical viscosity prevents the growth of the rayleigh - taylor instability , or their initial conditions are post - bounce and do not include an entropy gradient .
all calculations suffer from low - resolution , explaining why , except in a few cases , the rayleigh - taylor growth in the simulations is typically ten times longer than the analytic predictions .
if the rayleigh - taylor convection can quickly reset the entropy profile , other instabilities can develop@xcite .
it is this assumption that forms the basis for the standing accretion shock instability@xcite . in this scenario ,
entropy / vorticity perturbations are advected downward in the flow and couple with acoustic perturbations .
this instability can drive low - mode oscillations , again converting potential energy released near the base of the pns into kinetic energy at the edge of the stalled shock .
like the rayleigh - taylor instability , the energy in this explosion can be estimated by the maximum energy stored in the convective region .
all in all , scientists have focused on a few variants in instabilities : * rayleigh - taylor instability : driven by entropy gradients , high - entropy ( or `` hot '' ) bubbles will rise from the pns surface . low entropy material will find low points in the stalled shock , pile up , and form downstreams@xcite . *
rotationally stabilized rayleigh - taylor , also known as the solberg - hiland instability@xcite : if the specific angular momentum increases as one moves outward in radius ( typical for rotating massive star structures ) , the angular momentum profile will stabilize against convection .
this can produce asymmetric explosions@xcite .
* composition gradients and rayleigh - taylor : composition gradients , e.g. gradients in the electron fraction in the pns core can drive or stabilize against convection ( if it counters the entropy gradient , the relative diffusion of heat and composition can drive a doubly diffusive instability ) .
this instability is studied primarily in convection within the pns@xcite .
* acoustic / advective instability ( the standing accretion shock instability is a variant of this instability ) : this instability , coupling vorticity and pressure waves , is slower to grow than rayleigh - taylor but may develop and dominate in some scenarios@xcite . and a range of energy sources : * thermal energy in the pns : ( gravitational potential energy released in initial collapse ) .
the energy is radiated from the ns in the form of neutrinos@xcite .
* gravitational energy released from infalling material striking the pns . shocked material may drive some material to rise or neutrinos from the accreting material can transfer energy . * pns oscillations : the accretion , especially in acoustic / advective instabilities , can drive oscillations in the ns@xcite , although see@xcite . *
rotational energy : if the star is rotating rapidly , a disk can form around the pns . this rotational energy can drive a dynamo that generates strong magnetic fields to drive an explosion@xcite . the rotation period during the engine for even the fastest - spinning stars is typically too long ( 70 - 770ms ) to wind a strong magnetic field during the initial engine , but as the pns contracts , this engine becomes viable@xcite . as the convective region expands with the launch of the explosion , the amount of mass in this region that can absorb neutrinos from the hot pns decreases , effectively shutting off this source of energy .
pns oscillations can deposit energy longer , but even this energy source diminishes .
however , new energy sources develop after the collapse .
both post - explosion fallback and rotational energy have been invoked to drive explosions .
supernova `` fallback '' is material initially ejected in the supernova explosion that ultimately falls back onto the pns .
two major scenarios have been proposed to explain this fallback focusing on a 1-dimensional picture of supernova explosions .
the first argued that the ejecta decelerates as it plows through the imploding star .
it produces an explosion , but loses enough energy to fall back onto the pns@xcite .
alternatively , a reverse shock is produced when the supernova shock enters the hydrogen envelope ( with its shallow density gradient ) .
this reverse shock drives fallback@xcite .
the primary difference between these two scenarios is that the reverse shock scenario only produces fallback after a considerable delay ( the shock must first reach the hydrogen envelope and then the reverse shock must decelerate the outward moving material causing it to fall back ) .
the ejecta deceleration model@xcite tends to produce considerable fallback in the first 1 - 15s@xcite .
the amount and nature of then fallback in 1-dimensional simulations depends not just on the characteristics of the explosion ( progenitor star and explosion energy ) , but also on the numerical implementation of the explosion @xcite .
indeed , those models using piston - driven explosions are limited to the late - time reverse - shock fallback .
energy injected models have more fallback primarily in the first 1 - 15s after the launch of the explosion @xcite ( for a comparison of these two engines , see @xcite ) . in multi - dimensional simulations ,
fallback occurs even as the shock is launched as low - entropy material flows down through the hot rising shock .
this is seen both in late - time explosion calculations and ejecta supernova explosions , making it difficult to distinguish fallback from the supernova mechanism itself @xcite . for this paper
, we will call any material accreting onto the ns after the launch of the shock `` fallback '' . this fallback shocks when it hits the cooling pns and some of the material is re - ejected in plumes ( fig .
[ fig : fallback ] ) @xcite .
these plumes can make up a sizable fraction of the explosion energy , especially in those weak explosions that produce considerable fallback .
the amount of fallback ties the nature of the explosion engine directly to the remnant mass . we will discuss fallback in much greater detail in section [ sec :
bhformation ] . and there is a modest amount of angular momentum ( @xmath10 ) that provides a preferred ejection direction ( along the rotation axis ) . ]
rotating stars produce asymmetric explosions that tend to leave behind disks around the pns .
magnetic fields are capable of extracting this rotation energy to drive a second explosion . after the launch of the explosion , the pns contracts .
any rotation disk will contract , speeding its rotation period and within a few seconds a dynamo can create a strong magnetic field .
this magnetic field may produce jet - like explosions with a ns compact remnant@xcite .
the inner core of a massive star often couples to outer layers , causing its angular velocity to be on par with these outer layers .
this means the core s angular momentum is quite low .
if , however , the pns collapses to a bh and the inner @xmath11 accretes onto the compact remnant , the higher angular momentum material of the outer core can form a rapidly spinning accretion disk around the black hole .
it is this disk that has been invoked to form the collapsar grb engine @xcite .
the collapsar model argues that a magnetic dynamo in the disk ( or neutrinos in the disk ) produce a highly - relativistic jet , producing a gamma - ray burst .
this jet eventually disrupts the star , halting accretion and limiting the black hole mass . as the mass of the star increases
, a new instability can occur wherby pair - creation in the core causes the star to collapse and undergo explosive oxygen and silicon burning@xcite . for stars with masses above @xmath12 , a sufficiently large fraction of the center of the star becomes so hot that photodisentragion occurs before explosive burning reverses the implosion , accelerating the collapse@xcite .
these high entropy cores halt the infall before reaching nuclear densities , forming a proto - bh and producing a very different neutrino signal from lower mass stars and produce , at best , a disk wind outburst much dimmer than any supernova@xcite .
the proto - bh keeps accreting until an event horizon forms around it .
such proto - bh persists out to masses above @xmath13@xcite and the fate can only be altered if the core is rapidly rotating .
there are 3 basic ways a bh forms in stellar collapse . * direct collapse : in systems where all the explosion mechanisms discuss in section [ sec : snengine ] fail to revive the shock , the pns continues to accrete and ultimately becomes too massive to support itself , collapsing to form a bh . in these systems ,
the mass of the bh is the mass of the star at collapse .
fryer@xcite argued that stars above @xmath14 ( neglecting mass loss from winds and binary interactions ) collapse directly to bhs .
more recent estimates predict a lower limit but with large variations caused by stellar mass loss@xcite . *
fallback collapse : in systems where a weak explosion occurs , much of the material initially ejected at the launch of the explosion falls back , causing the pns to collapse to a bh @xcite .
this formation scenario can form a wide range of masses for low - mass bhs . *
proto - bh formation : very massive stars ( above @xmath15 ) produce a proto - bh that accretes mass until an event horizon forms , causing it to become a true black hole@xcite .
like the direct collapse stars , these stars form bh masses set to the stellar mass at collapse .
most `` normal '' supernovae have explosion energies between @xmath16 , only 1% of the total potential energy released in the collapse .
the convective engine provides a natural explanation for this 1% efficiency : because energy deposition drops as soon as the shock is launched , the explosion energy is set to the energy stored in the convective region prior to the launch of the shock ( see@xcite for more details ) . in this manner
, the explosion energy depends upon the ram pressure from the infalling stellar material ; a function of both the stellar progenitor and the time of explosion .
figure [ fig : snenergy ] shows the explosion energies as a function of explosion times for 3 progenitor masses ( assuming no mass loss ) .
the maximum energy stored in the convective engine is never more than a few times @xmath17 , explaining why the energy in normal supernovae are only 1% the total energy released .
these strong explosions have very little fallback ( less than a few tenths of a solar mass ) and the remnants are typically nss . )
when the star implodes .
the total energy stored is proportional to the ram pressure ( @xmath18 ) that , in turn , is set by the mass infall rate : @xmath19 where @xmath20 and @xmath21 are , respectively , the infall density and velocity and @xmath22 is the radius of the stalled shock . as the carbon / oxygen shell becomes the dominant infalling material ,
the pressure drops considerably .
if the convective engine has stored up @xmath17 of energy in the turbulent region , it will explode .
after this time , the maximum energy that can be stored in the convective region is well below the typical @xmath23 supernova energy ( see@xcite for more details ) . ]
although this naturally explains most supernovae , if the explosion is delayed ( taking more than @xmath24 after bounce ) , explosion energies are typically less than @xmath17 .
these weaker explosions lead to considerable fallback , producing a range of remnant masses from massive nss ( @xmath25 ) to bhs with masses nearly as massive as those produced through `` direct '' collapse .
supernovae with considerable fallback provide a unique probe into the supernova engine .
figure [ fig : fallbackmass ] shows the mass of the fallback material as a function of time for multiple explosion energies and 3 different stellar progenitor masses .
most of the fallback happens in the first 1 - 15s after the launch of the shockni , e.g. sn 2005bf@xcite . ] . above @xmath26
, the explosion energy must be very strong to avoid significant fallback .
for example , a @xmath27 ( 2 foe ) explosion for our binary @xmath28 progenitor has over @xmath29 of fallback in the first 15s . single star and @xmath30 and @xmath28 binary stars ( in the binary systems , the hydrogen is removed during a common envelope evolution phase , removing reducing the mass)@xcite .
note that the fallback begins within the first second and most occurs within the first 15s . for stars with masses above @xmath31
, there is considerable fallback ( more than @xmath32 ) for any explosion with normal ( 1 - 3foe ) explosion energies . ] the cumulative distribution of remnant masses for both delayed and rapid engine models for 3 different metallicities is shown in figure [ fig : nsmass ] .
in these models , 80 - 90% of all compact remnants are nss .
for the rapid explosions , 75 - 95% of these nss have masses below @xmath33 .
very few remnants are produced with masses between @xmath25 and @xmath34 .
for the delayed explosion , a more distributed range of masses exist . with these differences , accurate observations of the mass distribution will determine whether a rapid or delayed explosion dominates the supernova engine ( shedding light into the convective instability ) and whether a new energy source is needed beyond that available in the convective region . .
very few remnants are produced with masses between @xmath25 and @xmath34 .
for the delayed explosion , a more distributed range of masses exist.,title="fig : " ] . very few remnants are produced with masses between @xmath25 and @xmath34 .
for the delayed explosion , a more distributed range of masses exist.,title="fig : " ] most observations of compact remnants are focused on compact remnants in binaries .
binary observations are a biased sample , limited to those systems that remain bound after the formation of the compact binary .
the violent collapse of a stellar core can impart large velocities on the compact remnants . to understand these systems
, we must understand the `` kicks '' imparted on these remnants at birth .
a growing set of both ns and bh binaries seem to require that the compact remnant receive a burst of momentum at formation@xcite .
observations of pulsar proper motions suggest that the distribution of these kicks contains a sizable fraction of neutrons stars moving with velocities in excess of @xmath35@xcite .
although uncertainties persist in these measurements ( distance errors , issues such as biases in the sampling - systems with stronger kicks are more likely to escape the galaxy ) , these observations place fairly strong constraints on kicks received by nss .
much less is known about the bh kick distributions ; aside from a few binaries that suggest kicks are needed for some systems , we have very little information on bh kicks@xcite .
although there exist many theories to explain kicks , nearly all have trouble explaining those pulsars with velocities in excess of @xmath36 and although there is a preference among theorists towards kicks produced by asymmetries in the supernova ejecta , no kick mechanism can truly be ruled out or in .
most of the kick mechanisms invoke asymmetries in the supernova explosion whereby momentum conservation requires that if the ejecta is driven preferentially in one direction , the compact remnant must be directed in the opposite direction with equal momentum .
this means that both the mass and velocity of the ejecta asymmetry is critical . in this way ,
grb jets have difficulties driving kicks . with only @xmath37 in the jet mass , even if the jet is one - sided with a lorentz factor of 500 , the kick on a @xmath38 bh is only @xmath39 .
but a number of both ejecta and neutrino asymmetries could produce large kicks : * binary disruption : the sudden ejection of the supernova can disrupt a binary star system . if the supernova has a lot of ejecta and the binary is close , this scenario produces kicks up to @xmath40 , well below the bulk of the pulsar velocities@xcite .
the ejecta in a binary can also impart bulk velocities on the bound binaries , but these velocities are extremely low and can not explain the fastest binary systems@xcite . * propeller : asymmetric emission of radio waves for pulsars where the magnetic field is off - centered and inclined to the axis of rotation@xcite . *
growth of pre - collapse asymmetries : although it has been argued that asymmetries in the initial structure could drive strong kicks , realistic calculations show that only abnormally large asymmetries can produce kicks above a few hundred @xmath41@xcite . * low - mode rayleigh taylor : as the rayleigh - taylor instability develops , the lowest modes grow roughly to the size of the stalled shock ( fig .
[ fig : snsim ] ) .
such low modes can drive asymmetries in the supernova explosion ejecta , producing kicks@xcite .
this mechanism has received some success@xcite .
* advective / acoustic instabilities : low - mode instabilities are characteristic in this instability and several groups have studied the kicks from these instabilities**???**. * neutrino asymmetry from neutrinosphere turbulence : magnetoacoustic instabilities producing neutrino bubbles akin to the photon bubble instability produces asymmetric neutrino emission as the neutrino bubbles rise to the neutrinosphere@xcite .
* neutrino oscillations : if the core has extremely strong magnetic fields , neutrino oscillations coupled to those magnetic fields , can produce strong neutrino asymmetries@xcite .
unfortunately , none of these kick mechanisms are well - enough understood to then infer kick distributions for bhs .
but the kick placed on the bh can be categorized into 3 fates@xcite : * no kick : the propeller mechanism for all bhs , binary disruption and all ejecta explosions for direct bhs , neutrino - driven kicks may work in all bh formation scenarios . *
equal momentum kicks : binary and ejecta - driven kick mechanisms are likely to produce kicks with equal momenta to their ns counterparts , hence the kick velocity is roughly @xmath42 where @xmath43 , @xmath44 are the ns , bh masses respectively and @xmath45 is the ns velocity ( taken from the proper motion studies of pulsars . for neutrino driven mechanisms , this case may hold for both fallback and direct collapse bh scenarios .
* enhanced momentum kicks : for convective - instability kick mechanisms , asymmetries are strongest in delayed explosions , exactly those explosions that produce a lot of fallback .
these systems might get the strongest kicks . for bh kicks ,
a few scenarios are typically implemented into population studies of binary bh systems : bhs receive no kicks whatsoever ( there is a growing belief that this is disproved by observations - e.g.@xcite ) , all bhs receive kicks with conserved momentum : @xmath46 , direct bhs receive no kicks and bhs with fallback receive kicks ( either with conserved momentum , or in the enhanced case , with the same velocity as the pulsar observations ) . if one of these can be demonstrated absolutely , we would have constraints on the kick mechanism for both nss and bhs .
in addition , these kick distributions must be understood to tie theoretical mass distributions to observations of binary systems .
currently , the primary constraints on the mass distributions of compact remnants is limited to binaries containing nss and bhs ( e.g. x - ray binaries and binary pulsars ) .
early analyses of these binaries suggested that ns masses clustered with a very narrow mass range around @xmath47@xcite and bhs clustered with a broader range ( @xmath48 ) around @xmath49@xcite .
but as new systems have been added and the data on existing systems has become more refined , it has become clear that the distribution of nss is more spread out with neutron star masses ranging from @xmath50@xcite to @xmath51@xcite .
bh mass measurements are even more difficult to make , relying upon a complex combination of both observations of x - ray binaries and modeling of photometric and spectroscopic data .
although the bh mass range has widened with time , one aspect of the mass distribution continues to persist : the existence of a gap between the ns masses ( ending around @xmath25 ) and bh masses ( roughly @xmath34 and above)@xcite . however , bear in mind that the gap is statistical in nature and some bhs have error bars that fall within this gap , but even if this gap is not pristine , it is likely that there is a dearth of remnants with masses between @xmath52 .
the distribution of remnant masses , along with theory predictions assuming rapid or delayed explosions , is shown in figure [ fig : belczyn ] .
the top panel shows the observed distribution ( using the most - likely masses ) , clearly showing the observed mass gap .
the bottom panel shows the number distribution for 3 different models : rapid explosion , delayed explosion , and former model used by the startrack code@xcite .
only the rapid explosion model produces a gap , suggesting that the turbulent engine must grow quickly .
this result supports a rayleigh - taylor instability .
but this is not the end of the story .
the current paradox with this gap is that if it truly exists , it argues against many supernovae with a large amount of fallback .
the supernova energies produced by our two simplistic models at zero and solar metallicity are shown in figure [ fig : sne ] .
the rapid explosion model which produces a gap , does not produce any weak fallback supernovae .
but such supernovae seem to exist@xcite .
how can we have weak supernovae but no fallback ? barring a theoretical misunderstanding ( possible ) , one of the results , or more likely our interpretation of the results , is incorrect .
perhaps there are remnants with masses within the `` gap '' region , but they are sufficiently rare that they have yet to be observed ( requiring a bigger sample ) .
this would allow some weak supernovae .
the primary difficulty in using the compact remnant masses to constrain the supernova engine is the small sample statistics and potentially biased data of the x - ray binary population .
gravitational waves , with the potential to double the number of systems in a few years , can provide new insight into the compact remnant distribution .
unfortunately , although gravitational waves accurately measure the chirp mass , it is much more difficult to calculate the component star masses@xcite .
figure [ fig : gw ] shows the range of possible inferred masses for a series of binary mass pairs ( 4 different chirp masses ) . as with our x - ray binary systems , we will have to use large - sample statistics to demonstrate a `` gap '' in the compact remnant masses . unfortunately
, gravitational wave observations are limited to binary systems .
although it is likely that gravitational waves will determine whether the gap truly exists , it will be difficult to determine whether this gap only exists in binary systems .
in x - ray binaries and binary compact mergers ( gravitational wave detections ) , the inferred bh mass is the mass of the bh at formation plus the additional mass accreted in the binary .
measurements of solitary bhs will provide a more direct measurement of the bh mass distribution at formation .
in addition , binary systems may be a biased set ( e.g. , bhs that receive strong kicks , perhaps the lowest mass bh systems , have a lower probability of remaining bound ) .
the biases in bh binary formation are not well understood . a hubble space telescope program is ongoing to look for single bhs using microlensing .
wfirst has the potential to do these studies and increase our understanding of single bh masses .
the connection between supernovae and compact remnants places us in an ideal position to use observations of one to constrain the other .
the next decade should bring about many new discoveries in these fields .
70 colgate s. a. , white r. h. 1966 , apj , 143 , 626 herant , m. , benz , w. , hix , w. r. , fryer , c. l. , colgate s. a. 1994 , apj , 435 , 339 houck , j. c. and chevalier , r. a. 1991 , apj , 376 , 234 fryer , c. l. , benz , w. , herant , m. 1996 , apj , 460 , 801 bethe , h. a. , rev .
phys . , 62 , 801 fryer , c. l. and young , p. a. 2007 , apj , 659 , 1438 houck , j. c. and chevalier , r. a. 1992 , apj , 395 , 592 blondin , j. m. , mezzacappa , a. and demarino , c. 2003 , apj , 584 , 971 blondin , j. m. and mezzacappa , a. 2006 , apj , 642 , 401 miller , d. s. , wilson , j. r. , mayler , r. w. 1993 , apj , 415 , 278 colgate , s. a. , herant , m. , & benz , w. 2003 , phys .
227 , 157 endal , a.s . , and sofia , s. 1978 , apj , 220 , 279 fryer , c.l . , and heger , a. 2000 , apj , 541 , 1033 keil , w. , janka , h .-
t . , and mller , e. 1996 , apj , 473 , l111 dessart , l. , burrows , a. , livne , e. , ott , c.d . 2006 , apj , 645 , 534 foglizzo , t. , galletti , p. , scheck , l. , & janka , h .- t .
2007 , apj , 654 , 1006 zwicky , f. 1938 , apj , 88 , 522 burrows , a. , livne , e. , dessart , l. , ott , c. d. , murphy , j. 2006 , new astronomy , 50 , 487 yoshida , s. , ohnishi , n. , yamada , s. 2007 , apj , 665 , 1268 akiyama , s. , wheeler , j.c . ,
meier , d.l . , lichtenstadt , i. 2003 , apj , 584 , 954 wheeler , j.c . , akiyama , s. , & williams , p.t .
2005 , ap&ss , 298 , 3 dessart , l. , burrows , a. , livne , e. , ott , c.d .
2008 , apj , 673 , l43 colgate , s. a. 1971 , apj , 163 , 221 woosley , s. e. 1989 , ny acad .
ann . , 571 , 397 macfadyen , a. i. , woosley , s. e. , heger , a. 2001 , apj , 550 , 410 young , p. a. 2007 , apj , 664 , 1033 zhang , w. , woosley , s. e. , heger , a. 2008 , apj , 679 , 639 ugliano , m. , janka , h .- t . , marek , a. , & arcones , a. 2012 , apj , 757 , 69 dexter , j. , & kasen , d. 2013 , apj , 772 , 30 fryer , c. l. 2009 , apj , 699 , 409 woosley , s. e. 1993 , apj , 405 , 273 popham , r. , woosley , s. e. , fryer , c. l. 1999 , 518 , 356 barkat , z. , rakavy , g. , & sack , n. 1967 , phys .
lett . , 18 , 379 woosley , s. e. , & weaver , t. a. 1982 , in supernovae : a survey of current rsearch , ed . m. j. rees & r. j. stoneham ( dordrecht : reidel ) , 79 bond , j. r. , arnett , w. d. , & carr , b. j. 1984 , apj , 280 , 825 carr , b. j. , bond , j. r. , & arnett , w. d. 1984 , apj , 277 , 445 fryer , c. l. , woosley , s. e. , & heger , a. 2001 , apj , 550 , 372 fryer , c. l. , & heger , a. 2011 , astron .
nachrichten , 332 , 408 fryer , c. l. , belczynski , k. , wiktorowicz , g. , dominik , m. , kalogera , v. , holz , d. e. 2012 , apj , 749 , 91 maeda , k. , tanaka , m. , nomoto , k. , tominaga , n. kawabata , k. , mazzali , p. a. , umeda , h. , suzuki , t. , hattori , t. 2005 , apj , 666 , 1069 bruenn , s. w. , mezzacappa , a. , hix , w. r. , blondin , j. m. , marronetti , p. , messer , o. e. b. , dirk , c. j. , yoshida , s. , astro - ph/1002.4914 ellinger , c. i. , rockefeller , g. , fryer , c. l. , young , p. a. , park , s. , astro - ph/1305.4137 fryer , c.l . 1999 ,
apj , 522 , 413 heger , a. , fryer , c.l . ,
woosley , s.e . ,
langer , n. , & hartmann , d.h .
apj , 591 , 288 fryer , c. l. , & kalogera , v. 1998 , apj , 499 , 520 fryer , c. l. , burrows , a. , & benz , w. 1998 , apj , 496 , 333 gualandris , a. , colpi , m. , zwart , s. p. , & possenti , a. 2005 , apj , 618 , 845 willems , b. , henninger , m. , levin , t. , ivanova , n. , kalogera , v. , mcghee , k. , timmes , f.x . , & fryer , c.l .
2005 , apj , 625 , 324 fragos , t. , willems , b. , kalogera , v. , ivanova , n. , rockefeller , g. , fryer , c. l. , & young , p. a. 2009 , apj , 697 , 1057 cordes , j. m. , & chernoff , d. f. 1998 , apj , 505 , 315 arzoumanian , z. , chernoff , d. f. , & cordes , j. m. 2002 , apj , 568 , 289 gott , j. r. i. , gunn , j. e. , & ostriker , j. p. 1970 ,
apj , 160 , l91 harrison , e. r. , & tademaru , e. 1975 , apj , 201 , 447 fryer , c. l. 2004 , apj , 601 , l175 herant , m. 1995 , phys .
, 256 , 117 buras , r. , rampp , m. , janka , h .-
t . , & kifonidis , k. 2003 , phys .
, 90 , 1101 socrates , a. , blaes , o. , hungerford , a. , fryer , c. l. 2005 , apj , 632 , 531 kusenko , a. , & segr , g. 1996 , phys .
77 , 4872 kusenko , a. , & segr , g. 1997 , phys .
b , 396 , 197 barkovitch , m. , dolivo , j .- c . ,
montemayor , r. , & zanella , j. f. 2002 , phys .
d. , 66 , 123005 barkovitch , m. , dolivo , j .- c . ,
montemayor , r. , & zanella , j. f. 2004 , phys .
d. , 70 , 043005 fuller , g. m. , kusenko , a. , mocioiu , i. , & pascoli , s. 2003 , phys .
d , 68 , 103002 kusenko , a. 2004 , int . j. mod .
d 2004 , 13 , 2005 fryer , c. l. , & kusenko , a. 2005 , apj , 163 , 335 whalen , d. j. & fryer , c. l. 2012 , apj , 756 , l19 thorsett , s. e. & chakrabarty , d. 1999 , apj , 512 , 288 bailyn , c. d. , jain , r. k. , coppi , p. , orosz , j. a. 1998 , apj , 499 , 367 van kerkwijk , m.h . ,
van paradijs , j. , zuiderwijk , e. j. 1995 , a&a , 303 , 497 demorest , p.b . , penucci , t. , ransom , s. m. , roberts , m. s. e. , & hessles , j. w. t. 2010 , nature , 467 , 1081 ozel , f. , psaltis , d. , narayan , r. , & mcclintock , j. e. 2010 , apj , 725 , 1918 farr , w. m. , sravan , n. , & cantrell , a. et al .
2011 , apj , 741 , 103 belczynski , k. , wiktorowicz , g. , fryer , c. l. , holz , d. e. , & kalogera , v. 2012 , apj , 757 , 91 moriya , t. , tominaga , n. , tanaka , m. , nomoto , k. , sauer , d. n. , mazzali , p. a. , maeda , k. , suzuki , t. 2010 , apj , 719 , 1445 hannam , m. , brown , d. a. , fairhurst , s. , fryer , c. l. , harry , i. w. 2013 , apj , 766 , l14 | when a massive star ends its life , its core collapses , forming a neutron star or black hole and producing some of the most energetic explosions in the universe .
core - collapse supernovae and long - duration gamma - ray bursts are the violent signatures of compact remnant formation . as such , both fields are intertwined and , coupled with theory , observations of transients can help us better understand compact remnants just as neutron star and black hole observations can constrain the supernova and gamma - ray burst engine .
we review these ties in this paper . |
the search for supersymmetric particles is now an integral part of all current , as well as future , experimental programs at high - energy colliders . aside from many attractive features of supersymmetry ( susy )
@xcite , the driving force for these searches comes from the recognition that weak - scale susy , which is introduced to solve the gauge hierarchy problem , requires that the susy partners of the standard model ( sm ) particles must be accessible to experiments that probe the tev energy scale .
if this is the case , a large number of gluinos and squarks will be produced at future hadron colliders such as the cern lhc ( operated at @xmath21 tev with luminosity @xmath22 @xmath23s@xmath24 ) and very large hadron collider ( vlhc ; operated at @xmath25200 tev with @xmath22@xmath26 @xmath23s@xmath24 ) .
once produced , gluinos and squarks will subsequently decay to the lightest neutralino ( @xmath3 ) .
this means that such hadron colliders will become the `` neutralino factory '' , which is capable of producing up to @xmath1 @xmath3 s per year .
the @xmath3 is usually assumed to be the lightest supersymmetric particle ( lsp ) and thus stable if @xmath6-parity is conserved .
it will then escape the detector , resulting in the famous missing energy signature for susy @xcite .
however , the @xmath3 might not be altogether stable : if there exists another superparticle lighter than @xmath3 , such as the gravitino ( @xmath4 ) or axino ( @xmath5 ) , the @xmath3 will decay into , _
e.g. _ , @xmath27 @xcite or @xmath28 @xcite .
such a light gravitino naturally exists in gauge - mediated susy breaking ( gmsb ) models [ 512 ] as well as in a class of no - scale supergravity ( sugra ) models @xcite .
a light axino can also exist in sugra models with the peccei - quinn ( pq ) symmetry @xcite . as another possibility
, the violation of @xmath6-parity leads to the unstable @xmath3 even if it is the lsp .
the @xmath3 will then decay into such modes as @xmath29 , @xmath30 and @xmath31 @xcite .
if the @xmath3 decay takes place inside the detector , the resultant event topology would be very different from that in case of the stable @xmath3 .
the experimental signatures have recently been studied for the @xmath3 decay into @xmath27 [ 1625 ] and @xmath28 @xcite , motivated by the single @xmath32 event observed in the cdf experiment at the tevatron @xcite . for the cdf event interpreted as the signal of the @xmath3 decay ,
the inferred decay length @xmath7 is much below 1 m. however , subsequent reports on diphoton @xmath33 distribution observed in the cdf @xcite and d@xmath34 @xcite experiments , as well as the analysis of the lep data at @xmath35 gev @xcite , do not give any further evidence for the @xmath3 decay into photon(s ) with such short @xmath7 .
therefore , the possibility is still open for the @xmath3 decay into @xmath27 or @xmath28 occurring mostly outside the detector .
actually , such a slow decay of @xmath3 appears to be favored at least for relatively light @xmath3 : original gmsb models @xcite prefer relatively high susy breaking scale , @xmath36 gev @xcite , implying @xmath37 km for @xmath38 gev . in case of the @xmath18 decay
, the allowed range of the pq symmetry breaking scale , @xmath1 gev@xmath39 gev @xcite , leads to @xmath40 km for @xmath38 gev .
if the @xmath3 decay is slow and mostly occurs outside the detector , the signature observed in usual collider experiments will be indistinguishable from that of the stable @xmath3 . hence the signal of the @xmath3 decay
would be missed entirely . even if a significant part of @xmath3 s produced at hadron colliders decay into photon(s ) inside the detector , it might be hard to detect the signature with usual detectors , because a huge number of particles produced will make it difficult to identify a single photon not coming from the interaction point .
in addition , the detectors for the approved experiments at the lhc are not designed to measure the direction of each photon precisely .
therefore , in order to search for a slow decay of @xmath3 , we propose a dedicated experiment with the interaction point surrounded by a thick shield , by which most of the collision products are dumped .
the @xmath3 would go through the shield and decay in a long decay tunnel .
the decay product of @xmath3 can then be detected by a detector located at the end of the tunnel .
we show that the slow arrival time and the large off angle ( to the direction of the interaction point ) of the decay product will provide an unambiguous signal of slowly decaying @xmath3 s .
we also show that , in case of the @xmath3 decay into a photon and a light invisible particle such as @xmath4 and @xmath5 , the mass of @xmath3 can be determined by using the correlation between the energy and the arrival time of the decay photon .
furthermore , by reconstructing the decay kinematics with the estimated decay position , one can determine whether the final state is two - body or not .
the remainder of this paper is organized as follows .
we first show in section [ sec : lhc ] how the lhc can be the neutralino factory . in section [ sec :
model ] , we briefly review the theoretical models of the @xmath3 decay into a gravitino or an axino .
our strategy of the search for a slow decay of @xmath3 is then described in section [ sec : scheme ] . in section [ sec : mass ] , we illustrate the method of determining the mass of @xmath3 after the discovery of its decay .
section [ sec : vlhc ] is devoted to the discussion of the @xmath3 decay search at vlhc .
we also discuss in section [ sec : other ] other searches which are possible with our setup , _
i.e. _ , ( i ) other decay modes of @xmath3 , ( ii ) slow decays of any other longlived neutral or charged particles , and ( iii ) heavy stable charged particles . finally , we present our conclusions in section [ sec : conc ] .
a realistic design for the detector is investigated with monte carlo simulations in appendix [ sec : dete ] . in appendix
[ sec : munu ] , we estimate the rates of high - energy prompt neutrinos and muons , which can go through the iron shield and thus become the potential background .
to simulate the @xmath3 production and decay at the lhc , we first generate various susy processes in @xmath41 collisions at @xmath21 tev using isajet 7.13 @xcite with the cteq2l parton distribution functions @xcite . in most cases ,
the dominant products are the gluino ( @xmath42 ) and squark ( @xmath43 ) , which will then decay into @xmath3 s through @xmath44 and @xmath45 if @xmath46 , and @xmath47 and @xmath48 if @xmath49 .
if the gluino and squark are heavy ( @xmath50 tev ) , the production of gauginos ( charginos and neutralinos ) could become dominant . as an example
, we take @xmath51 , @xmath52 and @xmath53 , and assume the gaugino mass unification , which is known to roughly hold not only in sugra models but also in most of gmsb models .
figure [ fig : cross ] shows the total cross sections for the production of gluinos , squarks and gauginos , as well as the anticipated numbers of events per 100 fb@xmath24 , at the lhc as a function of @xmath54 .
we note that each susy event contains two @xmath3 s in the final state . also shown in this figure
are the corresponding values of @xmath55 , which results from the gaugino mass unification .
it is worth noting that the cross sections increase by a factor up to @xmath56 when the next - to - leading order qcd corrections are properly included @xcite .
for the discussion below , we select three representative gluino and neutralino masses : @xmath57 gev and @xmath58 gev ( case 1 ) ; @xmath59 gev and @xmath60 gev ( case
2 ) ; and @xmath61 gev and @xmath62 gev ( case 3 ) . figure [ fig : angle ] shows the angular distributions of @xmath3 s produced at the lhc for ( a ) case 1 , ( b ) case 2 , and ( c ) case 3 .
the integrated luminosity is taken to be 100 fb@xmath24 , which corresponds to the runtime of @xmath63 s with @xmath22 @xmath23s@xmath24 .
the total numbers of produced @xmath3 s are @xmath64 , @xmath65 and @xmath66 for case 1 , 2 and 3 , respectively .
it can be seen that most of @xmath3 s are produced in the forward direction .
more events can thus be detected if one installs a detector at a smaller angle @xmath67 with respect to the colliding proton beam .
we here choose @xmath68 ( or @xmath69 ) for the central axis of the detector by compromising the rate of the signal and that of the potential background ( high - energy prompt neutrinos and muons ) as discussed in appendix [ sec : munu ] .
figure [ fig : eneut ] shows the energy spectra of produced @xmath3 s pointing to the detector which covers @xmath70 and an elevation angle of @xmath71 .
we find that , in all three cases , the majority of @xmath3 s are produced with relatively low energies , _
i.e. _ , the energy spectra have a peak at @xmath72 to 2 .
this means that most of the produced @xmath3 s are not extremely relativistic , having @xmath73 .
this makes the time measurement of the decay products a powerful tool for the identification of slowly decaying @xmath3 s .
supersymmetry ( susy ) is widely acknowledged to be the best motivated extension of the sm .
this is not only because it provides a natural solution to the gauge hierarchy problem , but also there exist many pieces of indirect evidence , such as the correct prediction of @xmath74 , the convergence of the sm gauge couplings at very high energies , and the heaviness of the top quark as required by the radiative breaking of the electroweak gauge symmetry .
moreover , susy can provide a plausible candidate for cosmological dark matter .
these hints , as a whole , appear to constitute circumstantial evidence for susy .
susy , however , must be broken because no susy partners of ordinary particles have been discovered .
the breaking of susy is usually conceived to occur in a hidden sector at some very high energies , which will probably remain unaccessible to direct collider experiments at least in the near future .
thus the investigation into the origin of the susy breaking is of great importance since it could be the unique window to look into the world of such high energies , perhaps up to the planck scale .
two scenarios have been proposed on how the susy breaking is communicated from the hidden sector to the observable sector : in one scenario , as in supergravity ( sugra ) models @xcite , susy breaking is mediated by gravity . in this case , the susy breaking scale @xmath16 is necessarily of order @xmath75 gev to give the superparticle masses around the weak scale . the other is the gauge - mediated susy breaking ( gmsb ) scenario [ 512 ] , where susy is broken at a scale as low as @xmath76@xmath63 gev , and the gauge interactions act as the messengers of susy breaking . because of this relatively low value of @xmath16 , the mass degeneracy among squarks or sleptons , which results from the fact that gauge interactions are flavor blind , is hardly broken by the evolution to the weak scale , in contrast to sugra models .
this ensures sufficient suppression of flavor changing neutral current ( fcnc ) processes .
in addition , gmsb models may also provide a solution to the susy @xmath77 problem @xcite .
these two scenarios also lead to distinctive consequences concerning cosmological dark matter . in sugra models ,
the lightest neutralino ( @xmath3 ) is usually the lsp , and can be the dominant component of dark matter . on the other hand
, gmsb models naturally predict that the gravitino ( @xmath4 ) comes out to be the lsp .
the gravitino with mass @xmath78 kev is claimed to be a good candidate for warm dark matter @xcite .
we note that it is practically impossible to directly detect the gravitino dark matter in laboratories , in contrast to the neutralino dark matter , whose direct detection is being pursued in many experiments . in this way , for finding the future direction of particle physics and cosmology , it is of crucial importance to determine which scenario is realized in nature .
as will be shown later , our experiment to search for a slow decay of @xmath3 can probe a wide range of @xmath7 , which corresponds to @xmath16 of order @xmath76 to @xmath79 gev in case of the @xmath13 decay .
this covers much of the @xmath16 range presently considered in gmsb models .
the gravitino mass @xmath80 is related to @xmath16 ( on the condition of the vanishing cosmological constant ) by @xmath81 where @xmath82 gev is the reduced planck mass .
this leads to @xmath83 and 24 kev for @xmath84 and @xmath63 gev , respectively .
note that the gravitino heavier than @xmath2 kev overcloses the universe in the standard cosmology .
such a heavy gravitino , however , is still allowed to be the dark matter if the reheating temperature after the inflation is low enough @xcite . the longitudinal component of the gravitino , _
i.e. _ , the goldstino , couples to matter with weak ( not gravitational ) interaction strength proportional to @xmath85 . through this coupling
, the @xmath3 will decay to a photon and a gravitino with a partial width of @xcite @xmath86 where @xmath87 is the square of the photino admixture of the @xmath3 . if the @xmath3 is a pure bino , @xmath88 .
this partial width is most likely to be the dominant one if the @xmath3 is the next - to - lightest supersymmetric particle ( nlsp ) with not too small @xmath87 , resulting in the decay length @xmath89 for @xmath90 gev and @xmath91 gev , this decay length is much longer than a typical detector size , leading that most of the decays will occur outside the detector .
it has recently been pointed out @xcite that such large @xmath16 ( @xmath92 gev ) is expected in original gmsb models @xcite .
in addition , other gmsb models [ 911 ] tend to predict that susy breaking occurs at relatively high ( @xmath93 gev ) energies , although @xmath94@xmath0 gev can naturally fit in with a very recent model @xcite . equation ( [ eq : ctau_g ] ) also indicates that the measurement of both @xmath14 and @xmath7 will give direct information on @xmath16 .
the peccei - quinn ( pq ) symmetry was introduced to solve the strong @xmath77 problem @xcite .
this symmetry should be explicitly broken , leading to a light pseudoscalar , the axion ( @xmath95 ) , whose mass is related to the pq symmetry breaking scale @xmath19 by @xmath96 where @xmath97 is the qcd anomaly factor of the pq symmetry .
the axion is one of good candidates for dark matter owing to its feeble interaction with matter .
the allowed range of @xmath19 is confined to @xmath1@xmath98 gev @xcite from laboratory experiments as well as from astrophysical and cosmological considerations .
if both the pq symmetry and susy hold in nature , there should exist the fermionic superpartner of the axion , the axino ( @xmath5 ) .
the axino can remain light even after the susy breaking in sugra models @xcite .
in particular , the axino with mass @xmath99 kev could serve as warm dark matter @xcite as is the case for the gravitino .
if such a light axino exists , the @xmath3 can decay into @xmath28 . in case of the @xmath3 being a pure bino ,
the decay width can be written as @xcite @xmath100 which corresponds to the decay length @xmath101 the lower limit on the pq symmetry breaking scale , @xmath102 gev , implies @xmath40 km for @xmath38 gev .
we propose a completely new type of collider experiment , which is schematically shown in fig .
[ fig : scheme ] . gluinos and squarks ( and possibly gauginos ) are produced in @xmath41 collisions at the interaction point and promptly decay into @xmath3 s .
the @xmath3 s then enter the decay tunnel after going through the shield , by which most of the collision products are dumped . as a simple example
, we here consider an iron shield of 10 m thickness , and take the tunnel length to be @xmath103 m , although a much longer tunnel is certainly possible .
the acceptance of the tunnel is taken to be @xmath70 and an elevation angle of @xmath71 .
we choose this angle @xmath67 by compromising the rate of the signal and that of the potential background ( see appendix [ sec : munu ] ) .
two layers of the anti - coincidence hodoscopes possibly with track detectors are installed just on the back end of the shield for the rejection of punchthrough muons .
it can also be used for the trigger of heavy longlived charged particles . in addition
, the scintillation counters and/or track detectors can be installed on the wall around the decay tunnel to reject potential background events produced by cosmic rays . if the @xmath3 decays to a photon and a light invisible particle in the tunnel , the decay photon will reach the end of the tunnel and enter a shower detector which has a spherical front face , and initiate the electromagnetic shower . the incident position , angle , and energy of the photon , as well as the arrival time relative to the rf clock of the accelerator ,
are measured by the detector . instead of the bulk iron shield , we can use ( at extra costs ) an `` active '' @xmath104-shield , consisting of segmented calorimeters , magnetized iron plates and the muon trackers .
this setup will enable us to measure the total ( and missing ) transverse energy of the event and to tag the existence of high - energy muons .
the decay probability of @xmath3 within a distance of @xmath105 is given by @xmath106 where @xmath107 is the lorentz factor , and @xmath7 is the decay length . as a typical case
, the probability that the @xmath3 with @xmath108 decays in the tunnel ( 10 m @xmath109 51.6 m ) is calculated to be 20% for @xmath110 m , and 2.4% for @xmath111 m. in the following monte carlo simulations , we assume that the decay photon is emitted isotropically with energy equal to @xmath112 in the rest frame of the @xmath3 , as is the case for the @xmath13 decay with @xmath113 . the photon 4-momentum ( @xmath114 , @xmath115 ) in the laboratory frame is then calculated by the lorentz boost .
we show in fig .
[ fig : sample ] a typical generated event of the @xmath3 decay into a photon and a light invisible particle .
figure [ fig : neut ] shows the distributions of the lorentz factor @xmath116 for @xmath3 s entering the decay tunnel for case 1 , 2 and 3 in ( a ) , ( b ) and ( c ) , respectively .
the numbers of @xmath3 s entering the tunnel are @xmath117 , @xmath118 and @xmath119 for case 1 , 2 and 3 , respectively , for total integrated luminosity of 100 fb@xmath24 , which we assume for all the simulation plots shown in this paper .
the distributions of @xmath116 for @xmath3 s decaying in the tunnel are also shown in fig .
[ fig : neut ] .
we here take as an example the decay lengths calculated by eq .
( [ eq : ctau_g ] ) for @xmath120 gev , _
i.e. _ , @xmath121 km , 350 m , and 52 m for @xmath58 gev ( case 1 ) , 78 gev ( case 2 ) and 115 gev ( case 3 ) , respectively
. the resultant numbers of @xmath3 s decaying in the tunnel are @xmath122 , @xmath123 and @xmath124 for case 1 , 2 and 3 , respectively . also shown in this figure by hatched histograms
are the distributions for @xmath3 s whose decay photons enter the shower detector .
the corresponding energy spectra of the decay photons are shown in fig .
[ fig : gamma ] for ( a ) case 1 , ( b ) case 2 , and ( c ) case 3 .
the spectra have a peak at @xmath125 and extend up to @xmath15 gev . the clear separation of the @xmath3 decay signal from the potential background , _
i.e. _ , high - energy prompt neutrinos and muons as well as the cosmic - ray events , can be attained by measuring the arrival time and the off angle @xmath126 to the direction of the interaction point ( see fig .
[ fig : scheme ] ) .
figure [ fig : psitof ] shows the scatter plots of @xmath126 vs. the arrival time for ( a ) case 1 , ( b ) case 2 , ( c ) case 3 , and ( d ) background neutrinos and muons ( @xmath127 ) .
the background @xmath127 events all have @xmath128 , and arrive in narrow bunch structure ( here we assume @xmath129 ns ) .
on the other hand , most of the decay photons have large @xmath126 ( up to @xmath130 ) and arrive with significant delay ( typically 120 ns ) , because most of the produced @xmath3 s are not extremely relativistic .
we show in fig .
[ fig : tof ] the arrival time distributions .
the evidence for a slow decay of the heavy parent particles is clear in the peak position , which is displaced by about 2 ns from the exact bunch position , as well as in the long tail after the peak .
the decay length @xmath7 can in principle be estimated from the distribution of the events in the scatter plot of @xmath126 vs. the arrival time .
moreover , one could measure @xmath7 from the variation of the signal count rate correlated with the change of @xmath131 , which can be performed by moving the detector on rails .
together with the measured value of @xmath14 ( see section [ sec : mass ] ) , one can determine the strength of the interaction which governs the @xmath3 decay .
in particular , if one interprets the observed events as the @xmath13 ( @xmath18 ) decay , the susy breaking scale @xmath16 ( the pq symmetry breaking scale @xmath19 ) can be determined .
the large off angle @xmath126 and the slow arrival time are the generic features of slowly decaying heavy particles , and will thus always provide a clear signature of slowly decaying @xmath3 s . based on this prospect
, we now estimate the explorable range of the decay length @xmath7 .
figures [ fig : psitof ] and [ fig : tof ] represent typical examples of the off angle @xmath126 and the arrival time distributions for @xmath7 much longer than ( cases 1 and 2 ) or comparable to ( case 3 ) the tunnel length . for longer @xmath7 , the characteristic shape of the distribution will not change , while the number of detected events will decrease in proportion to @xmath132 . in order to estimate the longest explorable decay length
, we assume that 10 events with such a characteristic distribution are enough to discover the signal . for @xmath7 much shorter than the tunnel length , most of the detected events will be due to the @xmath3 decaying immediately after going through the shield .
such events will show the smaller off angle and shorter arrival time as compared to the cases shown in figs .
[ fig : psitof ] and [ fig : tof ] .
the off angle will nevertheless have finite non - zero values ( typically of 1 to @xmath133 ) , and the arrival time will definitely be deviated ( by 1 to 10 ns ) from that of background @xmath127 events , because the @xmath3 flies at least the shield thickness before decaying . in estimating the shortest explorable decay length ,
we assume 100 such events are enough to discover the signal .
figure [ fig : reach ] shows the resultant explorable range of @xmath7 for various @xmath14 at the lhc with 300 fb@xmath24 .
it can be seen that the explorable range of @xmath7 is wide , _
i.e. _ , 0.2 m to @xmath8 km for @xmath9 gev , and 1 m to 2 km for @xmath10 gev .
we also show in this figure the predicted curves of @xmath7 for the @xmath3 decay into @xmath27 and @xmath28 for typical values of @xmath16 and @xmath19 , respectively .
we find that one can explore @xmath16 in a wide range of @xmath134 to @xmath12 gev through the @xmath13 decay .
this range includes an interesting case of @xmath135 gev , where the gravitino has mass @xmath78 kev and can be the dominant component of dark matter in the standard cosmology . in case of the @xmath18 decay , @xmath19 can be explored up to @xmath136 gev for @xmath38 gev .
after discovering the signal of the @xmath3 decay , one can determine the mass of @xmath3 by using the correlation between the energy and the arrival time of the decay photon , if enough number of decay events can be accumulated . for this purpose , we first extract the approximate time - of - flight ( tof ) of @xmath3 , @xmath137 , from the time difference between the signal of the shower detector and the rf clock of the accelerator , as illustrated in fig .
[ fig : time ] . in this process , we assume that each @xmath3 comes from the closest possible bunch of the proton beam on the condition of @xmath138 .
the probability of attributing an event to a wrong starting bunch is small , _
i.e. _ , 2025% for the three cases described above . using the extracted value of @xmath137
, we define the parameter @xmath139 as @xmath140 where @xmath141 is the distance from the interaction point to the front face of the shower detector .
the distributions of @xmath142 are shown in fig .
[ fig : beta ] for ( a ) case 1 , ( b ) case 2 , ( c ) case 3 , and ( d ) background @xmath127 .
there is an upper cutoff at @xmath143 , which results from the modulation of @xmath137 by the bunch spacing of 25 ns .
the background @xmath127 events make a sharp peak at @xmath144 with no tail in contrast to the @xmath3 decay signal . figure [ fig : scan ] shows the scatter plots of @xmath114 vs. @xmath145 for ( a ) case 1 , ( b ) case 2 , and ( c ) case 3 .
the photon energy resolution is assumed to be @xmath146 .
the lower cutoff of @xmath147 at @xmath56 corresponds to the above - mentioned cutoff of @xmath142 at @xmath148 .
one can also find an edge with constant ratio of @xmath114 to @xmath147 .
this edge is composed of the events in which the decay occurs near the front face of the shower detector with the decay photon emitted toward the flight direction of the @xmath3 . for such events , @xmath141 and @xmath137 are nearly the same as the actual flight length and tof of the @xmath3 , respectively , and the decay photon has the full energy of the @xmath3 , _
i.e. _ , @xmath149 , resulting in the edge corresponding with @xmath150 . from the position of this edge
, the mass of @xmath3 can be determined with an accuracy of @xmath151% . for this purpose
, the shower detector should be designed to measure the photon energy in a wide range of 10 to 1000 gev .
figure [ fig : mass ] shows the corresponding distributions of @xmath152 for cases 1 to 3 .
an upper endpoint at @xmath150 can be seen in each case .
we now try to determine the nature of the @xmath3 decay , _ i.e. _ , whether it is two - body or not . although the symmetric peak at @xmath153 in fig .
[ fig : mass ] may be the first hint of the two - body decay , we can prove the two - body nature by directly reconstructing the decay kinematics .
for this purpose , we first select events with @xmath154 based on the @xmath126 distribution shown in fig .
[ fig : psi ] for ( a ) case 1 , ( b ) case 2 , ( c ) case 3 , and ( d ) background @xmath127 . after this
@xmath126 cut , only events in which the @xmath3 decays relatively near the shower detector can survive .
the efficiency of this cut is @xmath155 for each of the three cases .
figure [ fig : scanc ] shows the scatter plots of @xmath114 vs. @xmath147 for cases 1 to 3 after the cut of @xmath154 .
it can be seen that only photons with relatively low energies survive . thus the two methods of determining the @xmath3 mass , _
i.e. _ , the use of the edge in the scatter plot of @xmath114 vs. @xmath147 , and the direct reconstruction of the decay kinematics , are complementary in that different events are mainly used in these two methods .
we then simply assume that the decay position d is the middle point between c and e , both of which are defined in fig .
[ fig : scheme ] . using this assumption ,
the velocity of the @xmath3 can be approximately expressed by @xmath156 where @xmath157 and @xmath158 represent the flight path of the @xmath3 and that of the photon , respectively .
one can then calculate the reconstructed mass as @xmath159 where @xmath160 is the angle between the momentum of the @xmath3 and that of the photon ( see fig .
[ fig : scheme ] ) .
figure [ fig : recon ] shows the distributions of @xmath161 for ( a ) case 1 , ( b ) case 2 , and ( c ) case 3 .
it can be seen that the @xmath3 mass can be determined with an accuracy of @xmath162% for each case .
if one finds a peak in the distribution of @xmath161 at the right position expected from the edge in the scatter plot of @xmath114 vs. @xmath147 , one can conclude that the observed events are really the signature of the two - body decay of @xmath3 into a photon and an invisible particle .
very large hadron collider ( vlhc ) @xcite has been discussed recently as a post - lhc machine , being operated at @xmath25200 tev with @xmath22@xmath26 @xmath23s@xmath24 .
we stress here that the vlhc will become the neutralino factory much more powerful than the lhc .
figure [ fig : vcross ] shows the total cross sections for the production of gluinos , squarks and gauginos , as well as the anticipated numbers of events per 1000 fb@xmath24 , in @xmath41 collisions at @xmath25 tev as a function of @xmath54 . also shown in this figure
are the corresponding values of @xmath55 .
figure [ fig : vreach ] shows the explorable range of @xmath7 for the @xmath3 decay at the vlhc with 3000 fb@xmath24 . as in fig .
[ fig : reach ] , the upper end of the explorable region corresponds to the level of 10 decay events detected , and the lower one is derived from the condition that 100 decay events are detected .
the explorable range of @xmath7 is found to be 0.1 m to @xmath163 km for @xmath9 gev , 0.2 m to @xmath164 km for @xmath10 gev , and 1 m to 1 km for @xmath165 gev .
we also show in this figure the predicted curves of @xmath7 for the @xmath13 and @xmath18 decays for typical values of @xmath16 and @xmath19 , respectively .
it is found that one can extend the reach for @xmath16 up to @xmath166 gev through the @xmath13 decay . in case of the @xmath18 decay
, @xmath19 can be explored up to @xmath20 gev for relatively light @xmath3 .
it is worth noting that our experiment is unique in that one can explore such a high value of @xmath19 in laboratories .
in addition , our experiment can be performed under very high luminosity exceeding @xmath26 @xmath23s@xmath24 , in contrast to usual collider experiments .
one can search for other modes of the @xmath3 decay by using the setup proposed in this paper .
if the @xmath3 is sufficiently heavy , it can decay into @xmath167 or @xmath168 ( where @xmath169 is the higgs boson ) in the gravitino lsp scenario @xcite . in case
the @xmath3 is higgsino - like , the @xmath168 decay can even be dominant .
since the @xmath170 and @xmath169 bosons subsequently decay into @xmath171 or @xmath172 , the visible signature of these @xmath3 decays will be two hadron jets or two leptons without any signal in the anti - coincidence hodoscopes . with the help of track detectors at the end of the decay tunnel ,
the decay point can be exactly located , which makes the kinematical analysis of the decay and the mass determination much more streightforward than in case of the @xmath13 decay described in section [ sec : mass ] .
the @xmath3 can also decay into such modes as @xmath29 , @xmath30 and @xmath31 through weak violation of @xmath6-parity @xcite .
furthermore , the decay of other longlived neutral particles , _
e.g. _ heavy neutrinos , can be detected with the same setup and method .
our setup can also be used to search for slow decays of heavy longlived charged particles .
although the existence of such particles would in principle be indicated in collider experiments by charged particle tracks with ionization larger than the minimum one , the signature of their decay will be missed if it occurs mostly outside the detector . on the other hand , with our setup
, one can clearly detect the decay signal of such longlived particles as in case of the @xmath3 decay .
the `` anti - coincidence '' hodoscopes can be used as the trigger counter .
there are several candidates for longlived charged particles in susy models .
for example , if the gravitino or axino is the lsp , a charged superparticle such as the slepton ( @xmath173 ) can be the nlsp , and may decay into @xmath174 with @xmath7 much longer than the detector size .
the chargino ( @xmath175 ) can be another example of the longlived nlsp if the mass difference between the @xmath175 and the @xmath3 ( assumed to be the lsp ) is so small that it can decay only via @xmath176 @xcite .
weak violation of @xmath6-parity would also allow the existence of longlived charged particles @xcite .
in this case , the slepton can be the lsp , decaying slowly via , _
e.g. _ , @xmath177 .
furthermore , heavy longlived charged particles could also exist in the susy breaking sector and messenger sectors of gmsb models @xcite .
there are also many other speculations on heavy longlived charged particles [ 4649 ] .
if the @xmath7 of heavy charged particles is extremely long , they will appear to be `` stable '' particles even in our experiment .
one can search for such particles with our setup by precisely measuring their time - of - flight over 40 m distance between the `` anti - coincidence '' hodoscopes and the shower detector at the end of the tunnel . the @xmath178 distribution for heavy stable charged particles is expected to have a long tail following the peak at @xmath179 , in contrast to muons which make a sharp peak at @xmath180 .
long flight path in our setup leads to excellent resolution of @xmath181 , which will provide a clear signal for the heavy stable charged particle .
one could then determine its mass by measuring its momentum with magnetized iron and tracking chambers , which can be installed either in the active shield or at the very end of the shower detector .
we have investigated the detection of a slow decay of the lightest neutralino ( or any other longlived particles ) at the cern lhc and at vlhc . after pointing out that such hadron colliders will become the `` neutralino factory '' producing @xmath0@xmath1 @xmath3s / yr
if susy particles actually exist below @xmath2 tev , we have shown that a slow decay of @xmath3 can be detected in a dedicated experiment in which the collision products are dumped by a thick shield .
the decay product of @xmath3 can then be detected by a detector located at the end of a long decay tunnel .
the slow arrival time and the large off angle @xmath126 of the decay product will provide a clear signature of slowly decaying @xmath3 s . considering the @xmath3 decay into a photon and a light invisible particle such as the gravitino ( @xmath4 ) or axino ( @xmath5 ) , we find that one can explore the decay length ( @xmath7 ) of @xmath3 in a wide range , _
i.e. _ , 0.2 m to @xmath8 km for @xmath9 gev and 1 m to 2 km for @xmath10 gev for the case of the lhc .
this corresponds to the range of the susy breaking scale @xmath11 to @xmath12 gev in case of the @xmath13 decay . at vlhc
, one can extend the explorable range of @xmath14 up to @xmath15 gev , and that of @xmath16 up to @xmath17 gev . in case of the @xmath18 decay
, one can explore the pq symmetry breaking scale @xmath19 up to @xmath20 gev .
the mass of the decaying particle can be determined with an accuracy of @xmath151% using the correlation between the energy and the arrival time of the decay photon . with the setup we propose
, one can also search for ( i ) other decay modes of @xmath3 such as @xmath6-parity violating one , ( ii ) the decay of any other longlived neutral or charged particles , and ( iii ) heavy stable charged particles .
the experiment we proposed in this paper is a completely new type of collider experiment , which may turn out to be of crucial importance .
the civil engineering of the intersection region at future hadron colliders has to be designed so that it can accommodate long decay tunnels for such experiments .
all the calculations were performed on the rs/6000 workstations of the international center for elementary particle physics ( icepp ) , university of tokyo . k.m .
acknowledges a fellowship from the japan society for the promotion of science .
figure [ fig : dete ] shows a schematical view of a realistic design for the shower detector which we have investigated . to separate electromagnetic showers initiated by photons from the background , we design the detector consisting of the preshower detector ( ps ) , two parts of the electromagnetic calorimeter ( em1 and em2 ) , three planes of plastic scintillators ( s1 , s2 and s3 ) , the lead absorber ( l1 ) , and the hadron calorimeter ( had ) . in addition
, we can install the tracking chambers both in front of and behind the whole detector for the definite detection of high - energy muons ( and possibly heavy stable charged particles ) .
the preshower detector ( ps ) comprises two planes of multiwire proportional chambers ( mwpcs ) with cathode pad readout , with 2@xmath182 and 1@xmath182 lead converter in front , respectively ( radiation length @xmath182 being 0.56 cm for lead ) .
the scintillator s1 in front of ps can work as the anti - coincidence counter for charged particles .
the scintillator s2 , located just behind ps , is used for triggering .
the electromagnetic calorimeter is divided into two parts , each of which comprises modules with @xmath183 @xmath184 transverse dimensions .
the first part ( em1 ) , which is located at 170 cm downstream of ps , consists of 5 layers of lead ( 1@xmath182 ) and scintillators ( 1 cm ) .
the second part ( em2 ) , which is located at 220 cm downstream of em1 , consists of 20 layers of lead ( 1@xmath182 ) and scintillators ( 1 cm ) .
total radiation lengths of em1 and em2 are 5.1@xmath182 and 20.4@xmath182 , respectively .
the thick scintillator s3 , located between em1 and em2 , is used not only for triggering but also for the measurement of the arrival time .
the hadron calorimeter ( had ) contains 15 layers of 5 cm thick iron and 2 cm thick scintillators .
total thickness is @xmath185 nuclear interaction lengths .
hadronic showers will deposit much of its energy in had . between em2 and had , the lead absorber l1 of 20@xmath182 in thickness is inserted to totally absorb electromagnetic showers .
a coincidence of two planes of scintillators , _
i.e. _ s2 and s3 , constitutes the first trigger .
we also require the signals from em1 and em2 to exceed a threshold , which leads to the rejection of ( i ) muons going through the detector as minimum ionizing particles , and ( ii ) hadronic showers initiated by high - energy neutrinos in had . for this detector configuration ,
we have performed monte carlo simulations using geant 3.21 @xcite .
a typical event for 100 gev photon incident on the shower detector is shown in fig .
[ fig : event ] .
we can see that the electromagnetic shower starts to evolve in ps , reaches its maximum after passing em1 , and is mostly absorbed in em2 .
figure [ fig : pid ] shows the distributions of energy deposited in active layers of ( a ) s2 , ( b ) em1 , ( c ) em2 , and ( d ) had , by 100 gev photons and charged pions which are incident at @xmath186 on the shower detector .
it can be seen that electromagnetic showers ( initiated by photons ) can clearly be separated from hadronic showers ( initiated by pions ) using the deposit energy of each component .
the energy of incident photons can be determined from the sum of deposit energy in active layers of em1 and em2 , denoted as @xmath187 , whose distributions are shown in fig .
[ fig : eneres](a ) for photons incident at @xmath186 with energy @xmath188 , 50 , 100 and 200 gev .
the inefficiency concerning the scintillation light collection is not included .
figure [ fig : eneres](b ) shows the relation between @xmath189 and @xmath187 for @xmath186 and @xmath190 incidence .
it can be seen that good linearity is realized for @xmath188200 gev .
the resultant energy resolution of the shower detector as a function of incident photon energy is shown in fig .
[ fig : eneres](c ) .
the fit result for @xmath186 data is @xmath191 for @xmath188200 gev . the shower direction is obtained from the reconstructed position of the shower in ps and that in em2 . the position of the shower in ps can be determined from the pad signal of mwpcs , whose spatial resolution is assumed to be 5 mm .
the shower center in em2 can be estimated from the observed division of energy within a local cluster of modules .
we find that good spatial resolution ( @xmath192 cm ) in comparison with the size of the module ( @xmath183 @xmath184 ) can be obtained owing to the large gap between em1 and em2 , which makes the shower spread widely before entering em2 .
the resultant angular resolution averaged over various incident positions is shown in fig .
[ fig : angres ] .
it is found to be @xmath193 for photons with @xmath194200 gev .
the angular resolution could be improved if tracking chambers are used as active layers of em1 and em2 instead of scintillators .
although a large number of particles will be produced at the lhc , most of them ( hadrons , @xmath195 and @xmath116 ) will be dumped in our setup by the iron shield with thickness of @xmath196 nuclear interaction lengths .
however , high - energy neutrinos and muons can go through the shield and might interact with the material of the detector .
the main source of high - energy neutrinos and muons is expected to be `` prompt '' , _
i.e. _ , come from semileptonic decays of bottom and charm hadrons @xcite , while the contribution from the decay of @xmath197 and @xmath198 mesons is negligible at least in our setup .
in order to estimate the rates of these prompt neutrinos and muons , we have performed monte carlo simulations using isajet 7.13 @xcite for @xmath41 collisions at @xmath21 tev , by taking the total cross section of the bottom plus charm production to be 2 mb .
total @xmath199 events are generated , and the neutrinos and muons which point to the detector covering a polar angle of @xmath200 and an elevation angle of @xmath71 are recorded .
we consider four cases , _
i.e. _ , @xmath201 , @xmath202 , @xmath203 and @xmath190 .
the energy spectra of the prompt neutrinos are shown in fig .
[ fig : nu ] .
the spectra have a long high - energy tail when @xmath204 is small .
the neutrino could mimic the @xmath27 or @xmath28 signal if it interacts in the preshower detector and deposits enough energy ( see appendix [ sec : dete ] for the detailed design for the shower detector ) . as a simple estimate of such background
, we calculate the number of neutrinos which have energy @xmath205 gev and interact with @xmath206 g/@xmath184 lead .
the interaction cross section is taken to be @xmath207 @xmath184/nucleon .
the estimated numbers are @xmath208 , @xmath209 , @xmath210 and @xmath211 per 100 fb@xmath24 for @xmath201 , @xmath202 , @xmath203 and @xmath190 , respectively .
thus the background induced by the prompt neutrinos is negligibly small .
figure [ fig : mu ] shows the energy spectra of the prompt muons .
only the muons with @xmath212 gev can go through the shield because of the ionization energy loss .
the resultant rates of muons entering the tunnel are then estimated to be @xmath185 khz , @xmath213 hz , @xmath214 hz and @xmath215 hz for @xmath201 , @xmath202 , @xmath203 and @xmath190 , respectively , with luminosity of @xmath216 @xmath23s@xmath24 .
the single rate of the prompt muons will thus be comfortably small if we take @xmath217 .
high - energy muons can deposit a significant portion of their energy by bremsstrahlung or direct @xmath218 pair production , both of which could lead to the electromagnetic shower .
when combined with inefficiencies in the anti - coincidence counters , these muons might in principle mimic the @xmath27 or @xmath28 signal . folding the simulated spectra by the showering probability @xcite of muons , we estimate the numbers of muon - induced high - energy ( @xmath219 gev ) showers starting to evolve in the preshower detector to be @xmath220 , @xmath221 , @xmath222 and @xmath223 per 100 fb@xmath24 for @xmath201 , @xmath202 , @xmath203 and @xmath190 , respectively . assuming a veto inefficiency of @xmath224 by the staggered layers of anti - coincidence scintillators , the muon - induced background is expected to be negligibly small for @xmath217
. even if the rate of such muon - induced background turns out to be non - negligible , off - line analysis will eliminate such background based on the information on the arrival time as well as on the off angle @xmath126 .
99 for a review , see , _ e.g. _ , j.f .
gunion , report no . hep - ph/9704349 ( unpublished ) .
h. baer , c - h .
chen , f. paige and x. tata , phys .
d * 52 * , 2746 ( 1995 ) ; phys .
d * 53 * , 6241 ( 1996 ) .
n. cabibbo , g.r .
farrar and l. maiani , phys .
105b * , 155 ( 1981 ) .
nieves , phys . lett .
b * 174 * , 411 ( 1986 ) .
m. dine and a.e .
nelson , phys .
d * 48 * , 1277 ( 1993 ) ; m.
dine , a.e .
nelson and y. shirman , phys .
d * 51 * , 1362 ( 1995 ) ; m. dine , a.e .
nelson , y. nir and y. shirman , phys .
d * 53 * , 2658 ( 1996 ) .
m. dine , y. nir and y. shirman , phys .
d * 55 * , 1501 ( 1997 ) .
g. dvali , g.f .
giudice and a. pomarol , nucl . phys .
* b478 * , 31 ( 1996 ) ; a.e .
faraggi , phys . lett .
b * 387 * , 775 ( 1996 ) ; s.p .
martin , phys .
d * 55 * , 3177 ( 1997 ) ; s. dimopoulos and g.f .
giudice , phys .
lett b * 393 * , 72 ( 1997 ) .
a. de gouva , t. moroi and h. murayama , report no . hep - ph/9701244 ( unpublished ) .
t. hotta , izawa k .-
i . and t. yanagida , phys .
d * 55 * , 415 ( 1997 ) ; l. randall , report no .
hep - ph/9612426 ( unpublished ) ; n. haba , n. maru and t. matsuoka , report no .
hep - ph/9612468 ( unpublished ) ; hep - ph/9703250 ( unpublished ) ; y. shadmi , report no .
hep - ph/9703312 ( unpublished ) .
e. poppitz and s.p .
trivedi , phys .
d * 55 * , 5508 ( 1997 ) ; phys .
b * 401 * , 38 ( 1997 ) .
n. arkani - hamed , h. murayama and j. march - russell , report no .
hep - ph/9701286 ( unpublished ) ; h. murayama , report no .
hep - ph/9705271 ( unpublished ) .
i . , y. nomura , k. tobe and t. yanagida , report no .
hep - ph/9705228 ( unpublished ) .
j. ellis , k. enqvist and d.v .
nanopoulos , phys . lett .
* 147b * , 99 ( 1984 ) . t. goto and m. yamaguchi , phys .
b * 276 * , 103 ( 1992 ) ; e.j .
chun , j.e .
kim and h.p .
nilles , phys .
b * 287 * , 123 ( 1992 ) ; e.j .
chun and a. lukas , phys .
b * 357 * , 43 ( 1995 ) .
hall and m. suzuki , nucl . phys . * b231 * , 419 ( 1984 ) ; s. dawson , nucl .
b261 * , 297 ( 1985 ) ; r. barbieri and a. masiero , nucl . phys . *
b267 * , 679 ( 1986 ) ; h. dreiner and g.g .
ross , nucl .
b365 * , 597 ( 1991 ) .
k. kiers , j.n .
ng and g. wu , phys .
b * 381 * , 177 ( 1996 ) .
stump , m. wiest and c .- p .
yuan , phys .
d * 54 * , 1936 ( 1996 ) .
s. dimopoulos , m. dine , s. raby and s. thomas , phys . rev . lett . * 76 * , 3494 ( 1996 ) ; s. dimopoulos , s. thomas and j.d .
wells , phys .
d * 54 * , 3283 ( 1996 ) ; s. dimopoulos , m. dine , s. raby , s. thomas and j.d . wells , nucl .
. suppl . *
52a * , 38 ( 1997 ) .
s. ambrosanio , g.l .
kane , g.d .
kribs , s.p .
martin and s. mrenna , phys .
. lett . * 76 * , 3498 ( 1996 ) ; phys .
d * 54 * , 5395 ( 1996 ) ; s. ambrosanio , g.d .
kribs and s.p .
martin , report no .
hep - ph/9703211 ( unpublished ) . k.s .
babu , c. kolda and f. wilczek , phys .
77 * , 3070 ( 1996 ) .
bagger , k. matchev , d.m . pierce and r. zhang ,
d * 55 * , 3188 ( 1997 ) .
h. baer , m. brhlik , c - h .
chen and x. tata , phys .
d * 55 * , 4463 ( 1997 ) .
borzumati , report no .
hep - ph/9702307 ( unpublished ) .
dicus , b. dutta and s. nandi , phys .
lett . * 78 * , 3055 ( 1997 ) ; report no .
hep - ph/9704225 ( unpublished ) .
lopez and d.v .
nanopoulos , mod .
a * 11 * , 2473 ( 1996 ) ; phys .
d * 55 * , 4450 ( 1997 ) ; j.l .
lopez , d.v .
nanopoulos and a. zichichi , phys .
lett . * 77 * , 5168 ( 1996 ) ; report no .
hep - ph/9610235 ( unpublished ) ; phys .
d * 55 * , 5813 ( 1997 ) .
j. hisano , k. tobe and t. yanagida , phys .
d * 55 * , 411 ( 1997 ) .
s. park , in _ proceedings of the 10th topical workshop on proton - antiproton collider physics , fermilab , 1995 _ , edited by r. raja and j. yoh ( aip , new york , 1995 ) p. 62 .
d. toback , report no .
fermilab - conf-96/240-e ( unpublished ) .
s. abachi _ et al .
lett . * 78 * , 2070 ( 1997 ) .
j. ellis , j.l .
lopez and d.v .
nanopoulos , phys .
b * 394 * , 354 ( 1997 ) .
e.g. _ , e.w . kolb and m.s .
turner , _ the early universe _ ( addison - wesley , 1990 ) .
janka , w. keil , g. raffelt and d. seckel , phys .
lett . * 76 * , 2621 ( 1996 ) ; g. sigl , phys . rev . lett . * 76 * , 2625 ( 1996 ) .
h. baer , f.e .
paige , s.d .
protopopescu and x. tata , in _ proceedings of the workshop on physics at current accelerators and supercolliders _ , edited by j. hewett , a. white and d. zeppenfeld ( argonne national laboratory , argonne , 1993 ) .
j. botts _
et al . _ ,
b * 304 * , 159 ( 1993 ) ; h.l .
lai _ et al .
d * 51 * , 4763 ( 1995 ) .
w. beenakker , r. hpker , m. spira and p.m. zerwas , nucl .
b * 492 * , 51 ( 1997 ) .
e.g. _ , h.p .
nilles , phys .
110 * , 1 ( 1984 ) .
h. pagels and j.r .
primack , phys .
lett . * 48 * , 223 ( 1982 ) ; m. kawasaki , n. sugiyama and t. yanagida , report no .
hep - ph/9607273 ( unpublished ) .
s. borgani , a. masiero and m. yamaguchi , phys .
b * 386 * , 189 ( 1996 ) .
t. moroi , h. murayama and m. yamaguchi , phys .
b * 303 * , 289 ( 1993 ) .
peccei and h.r .
quinn , phys .
* 38 * , 1440 ( 1977 ) ; phys .
d * 16 * , 1791 ( 1977 ) .
k. rajagopal , m.s .
turner and f. wilczek , nucl .
b * 358 * , 447 ( 1991 ) .
d. denisov and s. keller , report no .
hep - ph/9701402 ( unpublished ) .
chen , m. drees and j.f .
gunion , phys .
d * 55 * , 330 ( 1997 ) .
krasnikov , phys .
b * 386 * , 161 ( 1996 ) .
s. dimopoulos , g.f .
giudice and a. pomarol , phys .
b * 389 * , 37 ( 1996 ) ; g. dvali and m. shifman , phys .
b * 399 * , 60 ( 1997 ) .
r. barbieri , g. dvali and a. strumia , nucl .
b391 * , 487 ( 1993 ) .
c.t . hill and e.a .
paschos , phys .
b * 241 * , 96 ( 1990 ) .
j. maalampi and m. roos , phys .
rep . * 186 * , 53 ( 1990 ) .
j.l . hewett and t.g .
rizzo , phys .
rep . * 183 * , 193 ( 1989 ) .
application software group , cern program library long writeup w5013 , 1993 ( unpublished ) .
atlas collaboration , atlas technical proposal , report no .
cern / lhcc/94 - 43 ( 1994 ) .
tsai , rev .
phys . * 46 * , 815 ( 1974 ) ; a. van ginneken , nucl .
instr . meth .
a * 251 * , 21 ( 1986 ) . | prospects are examined for the detection of a slow decay of the lightest neutralino ( or any other longlived particles ) at the cern lhc and at very large hadron collider ( vlhc ) .
we first point out that such hadron colliders will become the `` neutralino factory '' producing @xmath0@xmath1 neutralinos / yr , if gluinos and/or squarks actually exist below @xmath2 tev .
the lightest neutralino ( @xmath3 ) , usually assumed to be stable , will be unstable if lighter superparticles such as the gravitino ( @xmath4 ) or axino ( @xmath5 ) exist , or @xmath6-parity is not conserved .
the decay signal would , however , be missed in usual collider experiments , particularly when the decay mostly occurs outside the detector . in order to search for such a slow decay of @xmath3 ,
we propose a dedicated experiment where the collision products are dumped by a thick shield , which is followed by a long decay tunnel .
the decay product of @xmath3 can be detected by a detector located at the end of the tunnel .
the slow arrival time and the large off angle ( to the direction of the interaction point ) of the decay product will provide a clear signature of slowly decaying @xmath3 s .
one can explore the decay length ( @xmath7 ) in a wide range , _
i.e. _ , 0.2 m to @xmath8 km for @xmath9 gev and 1 m to 2 km for @xmath10 gev at the lhc .
this corresponds to the range of the susy breaking scale @xmath11 to @xmath12 gev in case of the @xmath13 decay predicted in gauge - mediated susy breaking models . at vlhc
, one can extend the explorable range of @xmath14 up to @xmath15 gev , and that of @xmath16 up to @xmath17 gev . in case of the @xmath18 decay
, the peccei - quinn symmetry breaking scale @xmath19 can be explored up to @xmath20 gev .
the mass of the decaying particle can be determined by using the correlation between the energy and the arrival time of the decay product . with the setup we propose
, one can also search for ( i ) other decay modes of @xmath3 such as @xmath6-parity violating one , ( ii ) slow decays of any other longlived neutral or charged particles , and ( iii ) heavy stable charged particles . `
ut - icepp 97 - 02 + hep - ph/9706382 + ` june 1997 * hadron colliders as the `` neutralino factory '' : + search for a slow decay of the lightest neutralino at the cern lhc * k. maki and s. orito _ department of physics , school of science , university of tokyo , tokyo 113 , japan _ 5.9 mm |
perhaps the most classical population model is the _ galton - watson branching process_. originally devised to model the survival of aristocratic patrilineal surnames , the galton - watson process may be described as follows : start with one existing particle . at every time step ,
all existing particles reproduce an independent number of offspring and die out .
the main question is then , what is the probability that the system survives forever ?
by use of generating functions it is fairly simple to analyze this model , and in fact it is well known that in a galton - watson process with offspring distribution @xmath1 , the probability of extinction is given by the unique minimal solution of the equation @xmath2 $ ] in the interval @xmath3 $ ] .
moreover , the solution @xmath4 satisfies @xmath5 if and only if @xmath6 \leq 1 $ ] .
( see _ e.g. _
@xcite for a thorough treatment . ) to make matters more interesting , one might add some geometry , by having the particles not only branch ( reproduce ) but also move in some underlying graph .
this is the _ branching random walk _ model , which is described as follows : start with one particle at some origin vertex @xmath7 in graph @xmath8 . at each time step , all existing particles reproduce an independent number of offspring and die out .
all offspring now independently choose a random neighbor of their parent s vertex , and move to that new position .
thus , a specific lineage of particles performs a random walk on @xmath8 . a different way to view
this model is as a tree - indexed random walk ( see @xcite for more on tree - indexed random walks ) where the domain tree is the tree of lineage formed by a galton - watson process .
see the pioneering work of biggins @xcite and the survey by shi @xcite .
both models mentioned above exhibit some sort of monotonicity , enabling coupling arguments . for example , put in an imprecise way , if one has more particles , the branching random walk is more likely to be recurrent .
the additional particles only help it return to the origin .
let us now introduce the model we work with , which we dub _ branching - annihilating random walk _ , or @xmath9 for short
. start with a single particle at some origin vertex @xmath7 of a graph @xmath8 . at each time step , all particles independently reproduce ( or branch ) into a random number of offspring .
these offspring then each choose independently a random neighbor of their parent s vertex and move to that neighbor .
( so far , everything is identical to the branching random walk . ) finally , at every vertex at which there is more than one particle , these particles are eliminated ( this is the annihilation phase ) .
@xmath9 is a model for population reproduction in some geometry , with a competition for resources .
the annihilation phase can be viewed as there being only enough resources for one particle at every vertex of the underlying graph .
let us stress that the difficulty in analyzing barw stems mainly from the lack of monotonicity .
adding particles may on the one hand assist in the ultimate survival of the system , but may also hinder the survival , as these additional particles may compete for resources and annihilate others , resulting in too few particles to survive .
it is most convenient to work with poisson distributed offspring , so for simplicity we will restrict to this distribution .
[ dfn : barw ] let @xmath10 be a real number .
let @xmath8 be a graph , and let @xmath11 be some vertex .
we define * branching - annihilating random walk * on @xmath8 , starting at @xmath7 , with parameter @xmath12 , or @xmath13 , as the following markov process on subsets of @xmath8 .
let @xmath14 be i.i.d .
poisson-@xmath12 random variables .
start with @xmath15 . for
every @xmath16 , given @xmath17 , define @xmath18 as follows .
suppose that @xmath19 .
for every @xmath20 , let @xmath21 be independent vertices chosen uniformly from the set @xmath22 ( the neighbors of @xmath23 in @xmath8 ) .
define @xmath24 by @xmath25 .
this is the number of offspring that have moved to @xmath26 .
finally , let @xmath27 . in the case
that @xmath28 then @xmath29 as well .
as remarked above , @xmath9 lacks monotonicity , and thus it is not easy to analyze .
however , it seems reasonable to ask the following immediate questions regarding the long term behavior .
some of these questions are being studied by the authors in a separate work , for the case of @xmath8 being the infinite @xmath30-regular tree .
suppose @xmath8 is an infinite transitive graph .
if @xmath12 is either too big or too small , one may dominate @xmath9 by a sub - critical galton - watson process .
thus , we are guaranteed extinction in either case .
( this is not surprising , as too little offspring do not give a good enough chance of survival , and too many offspring create too much annihilation , thus again ruining the chance of survival . ) the immediate questions that arise regard a super - critical interval of survival : * do there exist @xmath31 such that for @xmath32 there is positive probability of survival forever and for @xmath33 $ ] there is extinction a.s . ?
* if such an interval exists , what happens at the critical values @xmath34 and @xmath35 ? * can @xmath36 be identified ? in this paper we consider @xmath9 in the finite graph setting , and specifically on the complete graph .
of course , there is always a positive probability of extinction in one step on a finite graph , so on a finite graph @xmath9 will a.s .
die out at some finite time .
however , we may consider @xmath9 on a sequence of finite graphs with size tending to infinity , and try to understand asymptotic properties of the process for large graphs . in this work we consider the mean - field case , where the sequence under consideration is the complete graph on @xmath0 vertices as @xmath37 .
our first result states that @xmath9 on the complete graph has an exponentially large expected lifetime .
[ thm : extinction time ] for every @xmath38 there exists @xmath39 such that the following holds for all @xmath40 .
consider @xmath9 on the complete graph on @xmath0 vertices .
let @xmath41 be the extinction time of the process ( the first time there are no particles ) .
then , for each @xmath42 , @xmath43 \geq c e^{c n } .\ ] ] our main result regards the `` window '' of extinction .
it is not difficult to see that for @xmath9 on the complete graph on @xmath0 vertices , the number of particles will oscillate for a long time around the value @xmath44 .
our next result considers how long it takes the process to go extinct , once it has been conditioned to do so ; that is , how many steps did it take the process to reach @xmath45 particles , at the last excursion it made below the equilibrium point @xmath46 .
[ thm : main thm ] for every @xmath38 and @xmath47 , there exists @xmath48 such that the following holds for all @xmath40 .
consider @xmath9 on the complete graph on @xmath0 vertices and let @xmath49 be the number of particles at time @xmath50 .
let @xmath51 then for each @xmath52 , @xmath53 \leq c \log ( 1+x ) .\ ] ] though the above theorem holds for any @xmath38 , the conditioned chain @xmath54 exhibits remarkably different behaviors in two distinct regimes of the parameter @xmath12 - ( i ) @xmath12 is close to @xmath55 and ( ii ) @xmath12 is large ( see figure [ fig:1 ] and figure [ fig:2 ] ) .
our proof is general enough to tackle both regimes simultaneously
. it would be interesting to find out whether , for a fixed @xmath0 , the expected extinction time of the conditioned chain @xmath56 $ ] is decreasing with respect to @xmath12 .
@xmath9 , or rather a continuous time versions , have been studied before , see _
e.g. _ @xcite .
however , most focus on survival of the process , or stationary measures . on the other hand , recently there has been considerable interest among the physicists to study the behavior of a finite population evolving under some stochastic dynamics near its extinction time and particularly to find ` most probable or optimal path to extinction ' @xcite . to best of our knowledge
this is the first work to study the `` extinction window '' for @xmath9 ; that is , the length of the last path to extinction .
as our results show , at least in the mean - field case , this window is _ much _ smaller than the lifetime of the system , indicating that extinction is a `` catastrophic '' phenomenon - it occurs abruptly in a very short time frame .
our analysis makes heavy use of the fact that on the complete graph the geometry plays no role , so that @xmath9 can actually be seen as a markov chain on @xmath57 , making the model simpler .
it would be very interesting to understand the expected lifetime and extinction window in other finite graph settings .
let @xmath58 be a sequence of finite graphs with bounded degree and size growing to @xmath59 .
consider @xmath9 on @xmath60 with poisson-@xmath12 offspring .
* is it true that there exist critical @xmath31 such that if @xmath32 then the expected lifetime is exponentially large in @xmath61 , and if @xmath33 $ ] the expected lifetime is much smaller ( perhaps logarithmic ) ? * for which @xmath12 does @xmath9 on @xmath60 have a logarithmically small extinction window ? that is , for which @xmath12 does there exist small enough @xmath62 so that conditioned on extinction before reaching above @xmath63 particles , the conditioned process has logarithmically small expected lifetime ? the first and third authors are partially supported by grant no .
2010357 from the states - israel binational science foundation ( bsf ) .
this work was initiated when the second and third authors were in the statistical laboratory , university of cambridge .
thanks goes to nathanal berestycki for making this possible .
it will be much simpler to use the following equivalent form of @xmath9 on the complete graph on @xmath0 vertices .
( here is where the mean - field structure makes the analysis much simpler . ) given that @xmath64 , i.e. there are @xmath26 particles at time @xmath50 , every particle branches into poisson-@xmath12 particles , and each of these chooses a new vertex , independently , and uniformly among all @xmath0 vertices .
thus , due to the summability of the poisson distribution , at the branching phase every vertex receives an independent poisson-@xmath65 number of particles . in the annihilation phase
only those vertices with exactly one particle survive to the next step , which happens at a given vertex with probability @xmath66 .
thus , we have just shown that if @xmath67 is the number of existing particles in @xmath9 on the complete graph on @xmath0 vertices , then @xmath67 is a markov chain with transitions given by @xmath68 = \operatorname{\mathbb{p } } [ \mathrm{bin}(n , b(x ) ) = y ] = { n \choose y } b(x)^y ( 1-b(x))^{n - y } .\ ] ] this observation will be central it what follows .
we use the notation @xmath69 and @xmath70 to denote the probability measure and expectation of @xmath9 on the complete graph on @xmath0 vertices with ( poissonian ) offspring mean @xmath12 and with @xmath71 .
let @xmath72 .
consider the galton - watson process with offspring distribution @xmath73 .
it is well known that there exists a number @xmath74 such that the process dies out with probability @xmath4 , and that @xmath4 is the unique fixed point of @xmath75 = e^{-\lambda(1-s)}$ ] in @xmath76 .
also , since @xmath4 is the probability of extinction , it is clear that @xmath77 is a continuous strictly decreasing function of @xmath78 ( see _ e.g. _
@xcite ) . throughout
, we make extensive use of the following inequalities , which are easy to verify . * for any @xmath79 and @xmath80 , @xmath81 . * for any @xmath82
, we have @xmath83 . * for any @xmath82 , we have @xmath84 * the last two inequalities can be combined to deduce @xmath85 .
@xmath86 denotes the minimum of @xmath87 and @xmath88 denotes the maximum of @xmath87 .
we also make use of the stopping times @xmath89 since @xmath67 is non - negative , it is reasonable to write @xmath90 .
note that the @xmath91 signs are used to denote the hitting of below or above the corresponding subscript .
another tool we will use is the following standard large deviations result concerning binomial random variables . for @xmath92 , @xmath93 \leq \exp { \left ( - nb \cdot \tfrac{(1-\xi)^2}{4 } \right ) } .\ ] ] vs @xmath26 where @xmath94 $ ] for @xmath95 , @xmath96 and @xmath97 ( left ) and @xmath98 ( right ) .
note that for @xmath97 , @xmath99 is monotonically decreasing but @xmath100 is not linear , so @xmath99 can not be expressed as @xmath101 . on the other hand , for @xmath98 ,
the function @xmath99 is not even monotone - it first decreases and then it increases near @xmath102 .
[ fig:1 ] , title="fig : " ] vs @xmath26 where @xmath94 $ ] for @xmath95 , @xmath96 and @xmath97 ( left ) and @xmath98 ( right ) .
note that for @xmath97 , @xmath99 is monotonically decreasing but @xmath100 is not linear , so @xmath99 can not be expressed as @xmath101 . on the other hand , for @xmath98 ,
the function @xmath99 is not even monotone - it first decreases and then it increases near @xmath102 .
[ fig:1 ] , title="fig : " ] for @xmath95 , @xmath96 and @xmath97 ( left ) and @xmath98 ( right ) .
the probabilities are represented by colors - the blue represents high values and the red represents small values . for @xmath103 , from any @xmath26 in the tilted chain ,
the walker goes down by a multiplicative factor with high probability .
but for @xmath104 , the transition matrix is highly concentrated . for some @xmath26 ,
the tilted chain goes up with high probability . for some @xmath26
, it goes down with high probability . for a few @xmath26 s ,
the transition distribution is bimodal ! [ fig:2],title="fig : " ] for @xmath95 , @xmath96 and @xmath97 ( left ) and @xmath98 ( right ) .
the probabilities are represented by colors - the blue represents high values and the red represents small values . for @xmath103 , from any @xmath26 in the tilted chain , the walker goes down by a multiplicative factor with high probability .
but for @xmath104 , the transition matrix is highly concentrated . for some @xmath26 ,
the tilted chain goes up with high probability . for some @xmath26
, it goes down with high probability . for a few @xmath26 s ,
the transition distribution is bimodal ! [ fig:2],title="fig : " ]
in this section we prove theorem [ thm : extinction time ] .
let @xmath105 be the _ return _ time to one of the sites in @xmath106 $ ] , @xmath107 for the proof of theorem [ thm : extinction time ] we do not require the full strength of the following lemma , but it will also be required in the sequel . [
lem : sub and super ] let @xmath108 and small enough such that @xmath109 .
let @xmath110 , and define @xmath111 .
then @xmath112 where @xmath113 $ ] .
denote @xmath114 . on @xmath115 , we have that @xmath116
. so @xmath117 & = ( b(x)q_1 + 1-b(x))^n=(1-b(x)(1-q_1))^n \\ & \leq e^{-nb(x)(1-q_1 ) } = [ e^{-{\lambda}(1-q_1 ) e^{-\frac{{\lambda}x}{n}}}]^x \leq q_1^x , \end{aligned}\ ] ] where the last inequality follows by the definition of @xmath118 .
this implies that @xmath119 is a supermartingale .
we may apply the optional stopping theorem , @xmath120= { \mathbb{e}}[q_1^{x_{t_0 } } { \mathbbm{1}_{{\left\ { t_0<t_{{\varepsilon}n}^+ \right\ } } } } ] + { \mathbb{e}}[q_1^{x_{t_{{\varepsilon}n}^+ } } { \mathbbm{1}_{{\left\ { t_0>t_{{\varepsilon}n}^+ \right\ } } } } ] \geq g(x)+(1-g(x))q_1^n\ ] ] and therefore @xmath121 .
we obtain the lower bound similarly : @xmath122 & = ( b(x)q_2 + 1-b(x))^n=(1-b(x)(1-q_2))^n\end{aligned}\ ] ] now , @xmath123 , so a short calculation gives @xmath122 & \geq e^{-nb(x)(1-q_2)(1 + 2b(x)(1-q_2 ) ) } = [ e^{- { \lambda}e^{-\frac{{\lambda}x}{n } } ( 1-q_2)(1 + 2 b(x ) ( 1-q_2 ) ) } ] ^x\\ & \geq [ e^{- { \lambda}(1-q_2)(1 + 2{\lambda}{\varepsilon } ) } ] ^x = q_2^x .\end{aligned}\ ] ] this implies that @xmath124 is a submartingale . as before , by the optional stopping theorem , @xmath125= { \mathbb{e}}[q_2^{x_{t_0}}{\mathbbm{1}_{{\left\ { t_0<t_{{\varepsilon}n}^+ \right\ } } } } ] + { \mathbb{e}}[q_2^{x_{t_{{\varepsilon}n}^+}}{\mathbbm{1}_{{\left\ { t_0>t_{{\varepsilon}n}^+ \right\ } } } } ] \leq g(x)+(1-g(x))q_x^{{\varepsilon}n}\ ] ] and therefore @xmath126 . fix @xmath127 small enough so that : * it meets the requirements of lemma [ lem : sub and super ] . *
it satisfies @xmath128 , or equivalently , @xmath129 .
it follows that @xmath130 for all @xmath131 . *
it satisfies @xmath132 , or equivalently , @xmath133 . keeping in mind that @xmath134=0 $ ] for any @xmath135 , by the markov property we have that @xmath136 \leq \operatorname{\mathbb{p}}_x [ x_1 < { \varepsilon}n ] $ ] for all @xmath26 .
next , we bound the term @xmath137 $ ] using standard large deviations for the binomial distribution . note that by our choice of @xmath138 , for any @xmath131 we have that @xmath139=nb(x ) \geq nb({\varepsilon}n)\geq { \varepsilon}n \sqrt{{\lambda}}$ ] .
therefore , for any @xmath131 , @xmath140 \leq { \mathbb{p}}_x[x_1<{\varepsilon}n ] & \leq { \mathbb{p}}_x\big[x_1 < { \lambda}^{-1/2 } \cdot { \mathbb{e}}_x[x_1]\big ] \leq \exp(-c{\varepsilon}n ) \label{eqn : large down step}\end{aligned}\ ] ] where @xmath141 .
note that by lemma [ lem : sub and super ] we have that for all @xmath142 , @xmath143 \geq c':= 1 - q_1 ( 1-q_1)^{-1 } > 0 $ ] where @xmath144 .
thus , for @xmath145 we have that @xmath146 \geq c ' \cdot \inf_{y \geq { \varepsilon}n } \operatorname{\mathbb{e}}_y [ t_0 ] .\ ] ] so it remains to consider @xmath147 . by and the strong markov property , on the event @xmath148 , the random time @xmath41 dominates a geometric random variable with success probability @xmath149 .
thus , for all @xmath26 , @xmath146 \geq c ' \cdot e^{c { \varepsilon}n},\ ] ] which proves the theorem .
throughout this subsection we denote @xmath151 $ ] .
let @xmath152 such that the following inequalities hold : @xmath153 .
this is equivalent to @xmath154 which is possible since @xmath10 .
next , let @xmath155 be the transition function of our markov chain .
explicitly , for any @xmath156 , @xmath157 $ ] .
let @xmath158=nb(x)$ ] and @xmath159 .
[ lem : gamma ] for any @xmath160 , @xmath161 and @xmath162 , we have that @xmath163 where @xmath164 .
recall that @xmath165 .
the function @xmath166 is increasing for @xmath167 , which implies that @xmath168 is increasing while @xmath169 , and thus increasing as long as @xmath145 .
it now follows that @xmath170 is increasing in @xmath171 : @xmath172 and since @xmath173 , this expression is indeed increasing in @xmath171 .
it follows that @xmath174 so we want to bound from above the expression @xmath175 with a bound that is independent of @xmath26 . it can be simply checked that for @xmath176 we have @xmath177 so we want to bound @xmath178 . by the mean value theorem
, it will be sufficient to bound @xmath179 .
recall that @xmath180 , and @xmath181 , so @xmath182 is monotone increasing for @xmath183 . upon differentiation ,
we get for all @xmath184 $ ] , @xmath185 thus , @xmath186 [ lem : beta ] there exist constants @xmath187 and @xmath188 such that for any @xmath189 there exists @xmath190 such that for all @xmath191 , we have that @xmath192 recall that @xmath193 $ ] .
it follows immediately that @xmath194 for @xmath131 .
therefore , we only consider @xmath195 .
we have by the markov property , for @xmath196 , @xmath197 we bound the first term in using lemma [ lem : gamma ] .
@xmath198 for the second term , we use the upper bound of lemma [ lem : sub and super ] to obtain @xmath199 also by lemma [ lem : sub and super ] , @xmath200 note that as @xmath201 we have that @xmath202 and @xmath203 . combined with the assumption that @xmath204
, we can deduce that there exists @xmath205 such that @xmath206 is bounded away from @xmath55 uniformly in @xmath207 .
moreover , since @xmath208 , and since we assume that @xmath209 , we may take @xmath210 small enough so that for all @xmath211 we have @xmath212 . consequently , we can find @xmath213 large enough ( that depends only on @xmath210 ) such that @xmath214 plugging all this into , we conclude that there exist @xmath210 and @xmath215 and @xmath216 such that for all @xmath207 and for every @xmath217 , we have @xmath218 .
this proves the lemma for @xmath219 . as for @xmath220 , by lemma [ lem : sub and super ]
we have @xmath221 recall that @xmath222 so @xmath223 as @xmath201 . also , @xmath224 as @xmath225 and @xmath226 as @xmath201 . therefore , we may choose @xmath227 such that for all @xmath228 , @xmath229 .
thus , there exists @xmath230 such that if @xmath231 we have @xmath232 so we can take @xmath233 to obtain that @xmath234 and for all @xmath235 , sufficiently large @xmath0 and @xmath220 , we have @xmath236 .
wrap up by setting @xmath237 and @xmath238 .
( uniform lower bound ) .
[ lem : uniform lower bound ] fix @xmath72 .
there exists @xmath240 such that for all @xmath241 and @xmath242 , @xmath243 \geq \kappa \operatorname{\mathbb{p}}_x [ t_0 < t_u^+ ] .\ ] ] let @xmath244 , @xmath245 be two markov chains starting from @xmath246 respectively and with markov transition kernel @xmath247 .
consider the following coupling : * if @xmath248 then let @xmath249 , and given @xmath250 , @xmath251 * if @xmath252 then let @xmath253 , and given @xmath254 , @xmath255 next , given @xmath256 for @xmath257 , if @xmath258 then couple @xmath259 , and otherwise let @xmath260 evolve independently .
note that @xmath261 implies @xmath262 for all @xmath263 . by the mean value theorem , @xmath264 } \big| \tfrac{{\lambda}}{n}e^{-\tfrac{{\lambda}y}{n}}(1-\tfrac{{\lambda}y}{n})\big| \leq \tfrac{\lambda}{n } , \end{aligned}\ ] ] and since @xmath265 we get that @xmath266 .
we have , @xmath267}{{\mathbb{p}}[x\in { \left\ { t_0<t_{u}^+ \right\ } } ] } & \geq \frac{{\mathbb{p}}[y_1 = x_1 \ , \ x \in { \left\{t_0<t_{u}^+ \right\ } } ] } { { \mathbb{p}}[x\in { \left\{t_0<t_{u}^+ \right\}}]}\\ & = { \mathbb{p}}[y_1=x_1\big| x\in { \left\ { t_0<t_{u}^+ \right\ } } ] .\end{aligned}\ ] ] now , if @xmath268 then as @xmath37 , @xmath269 & \geq \sum_k \operatorname{\mathbb{p } } [ x_1 = k \big| x \in { \left\{t_0 < t_u^+\right\ } } ] \cdot \operatorname{\mathbb{p } } [ \mathrm{bin } ( n - k , \tfrac{b(x+1 ) - b(x ) } { 1-b(x ) } ) = 0 ] \\ & \geq \operatorname{\mathbb{p } } [ \mathrm{bin } ( n , \tfrac{e \lambda}{(e-1 ) n } ) = 0 ] \to e^{- e \lambda / ( e-1 ) } > 0 .
\end{aligned}\ ] ] similarly when @xmath270 , @xmath269 & \geq \operatorname{\mathbb{p } } [ \mathrm{bin } ( n , \tfrac{e \lambda}{(e-1 ) n } ) = 0 ] .
\end{aligned}\ ] ] we may take @xmath271 $ ] to complete the proof .
( geometric upper bound ) . [ lem : geometric upper bound ] fix @xmath72 and @xmath272 small .
then there exists @xmath273 such that for all @xmath274 , @xmath275 \leq\theta^x.\ ] ] consider the probability generating function of a @xmath276 random variable . since @xmath277 , this function has a unique nontrivial fixed point @xmath278 , satisfying @xmath279 ( here @xmath280 is the probability of extinction of a galton - watson process with poisson-@xmath281 offspring distribution ) . note that for any @xmath282 , we have @xmath283 , so @xmath284&=(1-b(x)(1-\theta))^n \leq e^{-nb(x)(1-\theta ) } \\ & = \big [ e^{-{\lambda}e^{-\frac{{\lambda}x}{n}}(1-\theta)}\big]^x \leq \big [ e^{-e^{{\lambda}{\varepsilon}}(1-\theta)}\big]^x = \theta^x .
\end{aligned}\ ] ] this implies @xmath285 is a supermartingale . since it is bounded , and @xmath286 is a.s .
finite , we may apply the optional stopping theorem to this supermartingale with @xmath71 for some @xmath287 .
we obtain @xmath288 = { \mathbb{e}}_x[\theta^{x_{t_0}}{\mathbbm{1}_{{\left\ { t_0<t_{u}^+ \right\ } } } } ] + { \mathbb{e}}_x[\theta^{x_{t_{u}^+}}{\mathbbm{1}_{{\left\ { t_0>t_{u}^+ \right\ } } } } ] \geq \operatorname{\mathbb{p}}_x [ t_0 < t_{u}^+ ] .
\end{aligned}\ ] ] we want to investigate what our process behaves like when conditioned on @xmath289 for @xmath290 and @xmath291 .
let @xmath292 $ ] .
we denote the transition matrix of the tilted chain as @xmath293 which is obtained by applying doob s h - transform to the original transition matrix @xmath247 w.r.t . the harmonic function @xmath294 .
the matrix @xmath295 is given by @xmath296= \frac{{\mathbb{p}}_x[x_1=y , t_0<t_{u}^+]}{{\mathbb{p}}_x|t_0<t_{u}^+ ] } = \frac{{\varphi}(y)p(x , y)}{{\varphi}(x)}.\ ] ] [ lem : st1 ] let @xmath241 and @xmath297 $ ] .
suppose @xmath298 for some @xmath299 and for all @xmath300 . then for any @xmath301 , the probability measure @xmath302 is stochastically dominated by the probability measure @xmath303 , where @xmath304 let @xmath305 .
we need to show that for any @xmath301 and @xmath306 , @xmath307\geq { \mathbb{p}}[z\leq k]$ ] , or equivalently , @xmath308 since @xmath309 is harmonic with respect to @xmath247 , this is equivalent to showing that @xmath310 so we write @xmath311 note that for @xmath312 , by the assumption , @xmath313 which implies that each term of the above sum is non - negative . [
lem : st3 ] let @xmath241 and @xmath297 $ ] .
suppose @xmath314 for some @xmath315 and for all @xmath316 .
then for any @xmath301 , the probability measure @xmath302 stochastically dominates the probability measure @xmath317 , where @xmath318 the proof is exactly similar to that of lemma [ lem : st1 ] where we replace ` @xmath59 ' in the bounds of the summands by @xmath319 .
we omit the details .
[ lem : st2 ] ( a ) fix @xmath320 .
let @xmath321 and @xmath322
. then @xmath323 .
\(b ) fix @xmath324 .
let @xmath325 and @xmath326 .
then for any @xmath327 , we have @xmath328 .
\(a ) we exhibit a coupling such that @xmath329 .
note that @xmath330 where @xmath331 are i.i.d.@xmath332 random variables , and @xmath333 where @xmath334 are i.i.d .
@xmath335 random variables .
so let @xmath334 be as such , and let @xmath336 .
it follows that @xmath337 and @xmath338 = \operatorname{\mathbb{p } } [ y_j = 0 ] = ( 1-p)$ ] , suggesting that indeed @xmath339 are i.i.d .
@xmath340 , and @xmath341 a.s .
\(b ) it suffices to show for any @xmath342 , @xmath343}{\operatorname{\mathbb{p } } [ x \le m ] } \ge\frac{\operatorname{\mathbb{p } } [ y \le k]}{\operatorname{\mathbb{p } } [ y \le m]},\ ] ] which , in turn , is implied by @xmath344 \operatorname{\mathbb{p } } [ y = j ] \ge \operatorname{\mathbb{p } } [ x = j ] \operatorname{\mathbb{p } } [ y = i ] $ ] for all @xmath345 . upon rearrangement of terms ,
the above is equivalent to @xmath346 which is obviously true .
[ lem : coupling1 ] denote the markov chain conditioned on @xmath347 by @xmath348 .
there exists @xmath349 and @xmath350 such that for all @xmath351 there exists @xmath230 such that for all @xmath352 the following holds . for any @xmath353
, we can couple @xmath354 with a sub - critical galton - watson process @xmath355 having offspring distribution @xmath356 , such that @xmath357 recall that the transition matrix of the conditioned chain @xmath358 is given by @xmath359 with @xmath360 $ ] .
it suffices to show that for any non - negative integers @xmath361 , @xmath302 is stochastically dominated by a @xmath362 random variable . by lemmas [
lem : beta ] and [ lem : st1 ] , we know that for small enough @xmath272 and large enough @xmath0 , @xmath302 is stochastically dominated by @xmath363
. fix @xmath195 .
note that @xmath364 which implies that @xmath363 is binomially distributed with @xmath0 trials and success probability @xmath365 that satisfies @xmath366 further , by lemma [ lem : st2](a ) , @xmath363 is stochastically dominated by a poisson random variable with mean @xmath367 .
note that @xmath368 .
so @xmath369 and thus @xmath370 .
also , one easily checks that @xmath371 for @xmath372 $ ] .
we thus obtain @xmath373 .
recall that @xmath374 is monotone on @xmath375 $ ] so for @xmath145 , we have @xmath376 .
hence , @xmath377 denote @xmath378 .
note that for @xmath201 we have that @xmath379 by lemma [ lem : beta ] .
so choose @xmath380 small enough so that @xmath381 .
putting all the ingredients together , we get that for all @xmath195 , @xmath382 and keeping in mind that for any two non - negative integers @xmath383 we have @xmath384 the proof is complete .
[ cor : extinction from small ] fix @xmath72 .
there exist @xmath349 such that for all @xmath351 there exists @xmath385 such that the following holds for all @xmath386 @xmath387 \leq c \log
( 1+x ) , \quad 0\leq x<{\varepsilon}n.\ ] ] [ lem : coupling2 ] denote the markov chain conditioned on @xmath388 by @xmath389 . given @xmath390 and @xmath391 , there exists @xmath392 such that the following holds . for any @xmath393
, we can couple @xmath394 with a subcritical galton - watson process @xmath395 having offspring distribution @xmath396 , such that with probability at least @xmath397 , @xmath398 where @xmath399 is as given in lemma [ lem : uniform lower bound ] with @xmath400 . by lemma [ lem : uniform lower bound ] and lemma [ lem : st3 ] , for any @xmath401 , the transition distribution @xmath402 stochastically dominates @xmath403 . in other words , @xmath404 where @xmath405 is distributed as @xmath406 , where @xmath407 .
by lemma [ lem : st2](b ) and from the simple inequality @xmath408 , we further have @xmath409 , where @xmath410 is distributed as @xmath411 .
clearly , @xmath412 where @xmath413 are i.i.d .
we can find @xmath415 such that for any @xmath386 and any @xmath416 , @xmath417 = ( 1 - \frac { \kappa x}{n } ) ^ { \lfloor \frac{n}{x } \rfloor } \le 1- { \underline{\gamma}}.\ ] ] consequently , @xmath413 stochastically dominates @xmath418 and hence @xmath419 . on the other hand , by hoeffding s inequality , @xmath420 \le \exp \big(-2(1- \kappa)^2 ( { \lambda}^{-1 } \log { \lambda}- { \varepsilon})^2n\big).\ ] ] thus , on the event @xmath421 , which happens with probability at least @xmath422 , the distribution @xmath402 stochastically dominates @xmath423 .
so , a simple union bound allows us to couple the conditioned chain @xmath424 with a subcritical galton - watson process having offspring distribution @xmath396 so that with probability at least @xmath425 , the subcritical galton - watson process is dominated by @xmath424 for the first @xmath426 steps .
this finishes the proof .
[ cor : lb ] given @xmath390 and @xmath391 , there exists @xmath385 such that the following holds for all @xmath386 , @xmath427 \geq c^{-1 } \log ( 1+x ) , \quad 0\leq x < { \mathrm{eq}}- { \varepsilon}n.\ ] ] by lemma [ lem : coupling2 ] , the conditioned chain @xmath424 can be coupled with the sub - critical galton - watson process @xmath428 with mean offspring @xmath429 such that @xmath430 and @xmath431 for all @xmath432 with probability at least @xmath433 .
let @xmath434 be the time of extinction for the process @xmath428 .
it follows from the standard theory of branching process that @xmath435 \ge c ' \log ( 1+x)$ ] for all @xmath436 . on the other hand , @xmath437 \le x \operatorname{\mathbb{e}}_1 [ v_t ] \le x { \underline{\gamma}}^t.\ ] ] by choosing @xmath438 with a sufficiently large constant @xmath439
, we obtain that @xmath440 \ge \frac{c'}{2 } \log ( 1+x)$ ] for all @xmath441 . by the above coupling
, @xmath427 \ge \operatorname{\mathbb{e}}_x [ s { \mathbbm{1}_{{\left\ { s \le d \log n \right\ } } } } ] - d \log n \cdot e^{-cn},\ ] ] which completes the proof .
throughout this section we set @xmath272 and @xmath442 $ ] . then , taking @xmath455 , by the strong markov property , for any @xmath444 , @xmath456 & \leq \sup_{x \geq \eta ' n } h(x )
\leq \theta^{r \eta n } < { \left ( \tfrac{\kappa}{2 } \right)}^{2 \eta n } \\ & \leq \inf_{x \leq \eta n } h(x)^2 \cdot 2^{-2 \eta n } . \end{aligned}\ ] ] note that since for @xmath457 , @xmath458 we have that if @xmath459 then @xmath460 \geq e^{\lambda { \varepsilon } } \cdot x_k$ ] .
let @xmath461 and let @xmath462
. by standard large deviations of binomial random variables , @xmath463 & \leq \operatorname{\mathbb{p}}_x [ x_1 \leq a^{-1 } \operatorname{\mathbb{e}}_x[x_1 ] ] \leq \exp { \left ( - \tfrac{(1 - a^{-1 } ) ^2}{4 } \cdot \operatorname{\mathbb{e}}_x [ x_1 ] \right ) } \\ & \leq \exp { \left ( - \tfrac{(a-1)^2}{4 } \cdot x \right ) } = p^x .\end{aligned}\ ] ] let @xmath464 be an integer such that @xmath465 .
then , @xmath466 & \geq \operatorname{\mathbb{p}}_x [ \forall \ 1 \leq j \leq m \wedge t_{ex}^+ , \ \
x_{j } \geq a x_{j-1 } ] \geq ( 1 - p^x ) ^{m } .\end{aligned}\ ] ] this holds for any @xmath26 .
thus inductively , for @xmath467 , @xmath468 & \geq \operatorname{\mathbb{p}}_x [ t_{ex}^+ \leq m ] \cdot \inf_{y \geq ex } \operatorname{\mathbb{p}}_y [ t_{{\mathrm{eq}}-{\varepsilon}n}^+ \leq ( k -1 ) m ] \\ & \ge \cdots \geq ( 1-p)^{m \log n } .\end{aligned}\ ] ] again this holds for all @xmath142 .
thus , @xmath469 is dominated by a geometric random variable of mean @xmath470 .
so we conclude that @xmath471 \leq ( km)^2 \cdot 2 ( 1-p)^{-2 \log n } , \ ] ] which is polynomial in @xmath0 as @xmath37 .
let @xmath472 denote the event @xmath473 . putting everything together
, we obtain that for @xmath474 , @xmath475 \\ & = \operatorname{\mathbb{e}}_x [ t_0 { \mathbbm{1 } _ { { \mathcal{e}}(\eta ' n ) } } \
| \ { \mathcal{e}}({\mathrm{eq}}-{\varepsilon}n ) ] + \operatorname{\mathbb{e}}_x [ t_0 \wedge t_{{\mathrm{eq}}-{\varepsilon}n}^+ \cdot { \mathbbm{1}_{{\left\ { t_{\eta ' n}^+ < t_0 < t_{{\mathrm{eq}}-{\varepsilon}n}^+ \right\ } } } } \
| \ { \mathcal{e}}({\mathrm{eq}}-{\varepsilon}n ) ] \\ & \leq \operatorname{\mathbb{e}}_x [ t_0 \ | \ { \mathcal{e}}(\eta'n ) ] \cdot \frac{\operatorname{\mathbb{p}}_x [ { \mathcal{e}}(\eta ' n ) ] } { \operatorname{\mathbb{p}}_x [ { \mathcal{e}}({\mathrm{eq}}-{\varepsilon}n ) ] } + \frac { \sqrt { \operatorname{\mathbb{e}}_x [ ( t_0 \wedge t_{{\mathrm{eq}}-{\varepsilon}n}^+ ) ^2 ] \cdot \operatorname{\mathbb{p}}_x [ t_{\eta ' n } < t_0 < t_{{\mathrm{eq}}-{\varepsilon}n}^+ ] } } { h(x ) } \\ & \leq c \log ( 1+x ) + km\sqrt{2 } ( 1-p)^{-\log n } \cdot
2^{-\eta n } \leq c ' \log ( 1+x ) , \end{aligned}\ ] ] for some constant @xmath476 .
[ lem : petunia ] given @xmath72 and @xmath477 , there exists a constant @xmath478 such that for all @xmath479 , @xmath480\leq c,\ ] ] where @xmath481 is the total time spent by @xmath482 in the interval @xmath483 .
we start with the observation that for @xmath484 , since @xmath485 , @xmath486 choose @xmath487 large enough such that @xmath488 . call a step @xmath489 of the markov chain @xmath482 _ unusual _ if @xmath490 and @xmath491 . by standard large deviations of binomial random variables , for @xmath92 , @xmath492 \leq \exp { \left ( - nb \cdot \tfrac{(1-\xi)^2}{4 } \right)}$ ] .
since @xmath493 , the probability of step @xmath489 being unusual is bounded by @xmath494 \leq \exp { \left(- ( e^{\lambda { \varepsilon}/2}-1)^2 \cdot \tfrac{\delta
n } { 4 } \right ) } .\ ] ] note that if @xmath495 and for all @xmath496 the step @xmath489 is _ not _ unusual , then by our choice of @xmath497 , @xmath498 that is , if all steps @xmath499 are not unusual then @xmath500 .
thus , @xmath501 implies that every time @xmath502 that @xmath503 , we must have that there exists @xmath504 such that @xmath489 is an unusual step . in conclusion , for any @xmath484 , @xmath505 & \leq { \mathbb{p}}_x[\text{$x$ takes at least $ \lfloor d / m \rfloor$ unusual steps } ] \\ & \leq \exp { \left ( - ( e^{\lambda { \varepsilon}/2}-1)^2 \cdot \tfrac{\delta}{4 } \cdot \lfloor \tfrac{d}{m } \rfloor \cdot n \right ) } .\end{aligned}\ ] ] on the other hand , by lemma [ lem : uniform lower bound ] , @xmath506 . combining the above two observations
, we obtain that for any @xmath484 , @xmath507 \le e^ { ( k_1 - dk_2)n}\ ] ] for constants @xmath508 and @xmath509 which are functions of @xmath510 and @xmath12
. now the assertion of the lemma follows immediately from the representation @xmath511 = \sum_{d=0}^\infty { \mathbb{p}}_x [ h \ge d\ | \ t_0<t_{{\mathrm{eq}}-{\varepsilon}n}^+].\ ] ] let us now prove the upper bound .
let @xmath62 be as in lemma [ lem : starting from small ] . for @xmath512 , theorem
[ thm : main thm ] follows directly from lemma [ lem : starting from small ] . for @xmath513 , by the strong markov property , @xmath514 \leq \operatorname{\mathbb{e}}_x [ t_{\eta n}^- \ | \
t_0 < t_{{\mathrm{eq}}-{\varepsilon}n } ] + \sup_{y < \eta n } \operatorname{\mathbb{e}}_y [ t_0 \ | \
t_0 < t_{{\mathrm{eq}}-{\varepsilon}n}^+ ] .\ ] ] we can bound @xmath515 \le 1 + \operatorname{\mathbb{e}}_x [ h \ | \
t_0 < t_{{\mathrm{eq}}-{\varepsilon}n } ] $ ] where @xmath516 and hence by lemma [ lem : petunia ] , @xmath515 \le c_1 $ ] . | we study a model of growing population that competes for resources . at each time step , all existing particles reproduce and the offspring randomly move to neighboring sites . then at any site with more than one offspring the particles are annihilated .
this is a non - monotone model , which makes the analysis more difficult .
we consider the _ extinction window _ of this model in the finite mean - field case , where there are @xmath0 sites but movement is allowed to any site ( the complete graph ) .
we show that although the system survives for exponential time , the extinction window is logarithmic . |
when transforming results from lattice simulations into a continuum scheme such as @xmath0 it is often desirable to know the perturbative expansion of the renormalized coupling in powers of the bare lattice coupling .
this is useful as an intermediate step for relating the @xmath0 coupling to the coupling defined in nonperturbative lattice schemes such as the ones based on the static quark potential @xcite and schrdinger functional @xcite , and is also needed to translate bare lattice quark masses into the @xmath0 scheme ( see , e.g. , @xcite ) .
the one loop coefficient in the expansion is of further interest because it determines the ratio of the lattice and @xmath0 @xmath1-parameters @xcite .
moreover , the one loop coefficient is also needed for determining the two loop relation between the couplings , from which the third term in the lattice beta - function ( governing the approach to the continuum limit ) can be determined @xcite . in this paper
we derive , for general lattice fermion formulation , a compact general integral formula from which the fermionic contribution to the one - loop coefficient in the perturbative expansion of the @xmath0 coupling in powers of the bare lattice coupling can be extracted .
the motivations for pursuing this are as follows .
first , given the plethora of lattice fermion actions currently in use , and the likelyhood of new ones or improved versions of present ones being developed in the future , it is desirable where possible to have general formulae from which quantities of interest can be calculated without having to do the calculation from scratch each time .
second , it is desirable to have independent ways to check the computer programs used these days to perform lattice perturbation theory calculations via symbolic manipulations . third , by reducing the calculation to a managable number of one loop lattice integrals one can more easily achieve greater numerical precision than with symbolic computer programs .
this is important , since , as emphasized in @xcite , the one loop results need to be determined with very high precision to achieve reasonable precision in the two loop result . as a demonstration that the general formulae of this paper are useful in this regard , we apply them to obtain the fermionic contribution to the one loop coefficient in the case of wilson clover fermions @xcite to almost twice as many significant decimal places as in the previous literature . as reviewed in sect .
ii , determining the fermionic contribution to the one loop coefficient reduces to determining a constant @xmath2 arising in a logarithmically divergent one fermion loop lattice feynman integral @xmath3 , which has the general structure i(am)=(a^2m^2)+c_i [ 1.1 ] here @xmath4 is the lattice spacing and @xmath5 an infrared regulator fermion mass .
the numerical factor in the log term is universal , whereas @xmath2 depends on the details of the lattice fermion formulation .
@xmath3 arises from the one fermion loop contribution to the gluonic 2-point function , and it is from this that it was evaluated in previous works for specific lattice fermion formulations . however , ward identities allow @xmath3 to also be evaluated from the gluonic 3- or 4-point functions . in this paper
we evaluate @xmath3 from the one fermion loop contribution to the gluonic 4-point function . in this case
there are five lattice feynman diagrams to consider rather than the two diagrams for the gluonic 2-point function
see fig .
[ vac1 ] .
@xmath6 nevertheless , evaluation of @xmath3 from the 4-point function turns out to be advantageous .
the diagrams are evaluated at vanishing external momenta without the need to first take momentum derivatives , and we find three nice properties : ( i ) only one of the five diagrams is logarithmically divergent it is the first @xmath7 diagram in fig . [ vac1 ] .
the other four diagrams are all convergent .
( ii ) the logarithmically divergent diagram is not affected by changes in how the link variables are coupled to the fermions ( e.g. , it is unchanged by adding staples , clover term etc . ) . consequently , it is the same for improved and unimproved versions of the lattice fermion formulation ( provided the free field formulations are the same ) .
( iii ) the four convergent diagrams , or subsets of them , vanish when the lattice dirac operator is sufficiently simple .
in particular , they all vanish for unimproved wilson and staggered fermions , also when the naik term @xcite is included .
thus for improved versions of wilson and staggered fermions the only new quantities to compute relative to the unimproved case are the four convergent one - loop lattice integrals .
the main result in this paper is a general integral formula for @xmath3 obtained by evaluating the contributions from the five @xmath7 feynman diagrams in fig .
[ vac1 ] for general lattice fermion formulation , from which the desired constant @xmath2 can be extracted .
specifically , we do the following : ( a ) evaluate the contribution from the logarithmically divergent diagram , deriving a quite explicit general formula which is seen to reproduce previous results for the cases of unimproved wilson and naive / staggered fermions , and ( b ) derive formulae for , and describe a straightforward procedure for evaluating , the contributions from the four convergent diagrams .
we illustrate this in the case of wilson clover fermions .
the general formulae lead to integrals to which the method of ref .
@xcite can be applied , reducing the integrals to basic lattice integrals that are already known to high precision .
the application of our result to other lattice fermion formulations such as asqtad staggered fermions @xcite and overlap fermions @xcite will be made in future work .
the paper is organized as follows .
ii reviews the one loop expansion of the @xmath0 coupling in the bare lattice coupling , using the background field approach . in sect .
iii we derive an initial expression for @xmath3 as the sum of contributions from the five @xmath7 diagrams in fig . 1 and point out the properties ( i ) , ( ii ) and ( iii ) mentioned above .
rather than evaluating the diagrams directly , we infer them from perturbative expansion of the fermion determinant , which is easier . from this
the general formulae and applications mentioned in ( a ) and ( b ) above are derived in sect .
iv and v , respectively . the concluding sect .
vi describes applications to be carried out in future work , as well as the possibility of deriving similar results for the gauge ghost contribution to the one loop coefficient .
also , it is pointed out that the present results are relevant for a previous proposal for constructing the gauge field action on the lattice from the lattice fermion determinant @xcite .
specifically , our results give an expression for the coefficient of the yang - mills action that arises in that proposal .
some technical details of our calculations are provided in two appendices .
the gauge field quantum effective action @xmath8 in euclidean spacetime can be expressed ( prior to gauge fixing ) as e^-(a)=q| e^-s_ym(a+gq)-|d(a+gq ) [ 2.1 ] where @xmath9 is the yang - mills action and @xmath10 is the dirac operator coupled to @xmath11 where @xmath12 is the background gauge field and @xmath13 the quantum fluctuation field .
the gauge group is taken to be su(n ) with the fermion fields being in the fundamental representation .
@xmath8 has the loop expansion ( a)=s_ym(a)+_1(a)+g^2_2(a)+ +g^2(n-1)_n(a)+ [2.2 ] expanding @xmath8 in powers of @xmath12 , the term quadratic in @xmath12 has the form ^(2)(a)=_p^(2)(p)_^ab_^a(p)_^b(-p ) [ 2.3 ] where @xmath14 denotes the fourier transform of @xmath15 .
gauge invariance of @xmath8 and rotation symmetry imply that the 2-point function has the form ^(2)(p)_^ab=-^ab(p^2_-p_p _ ) -_1(p)- -g^2(n-1)_n(p)- [2.4 ] the relation between the @xmath0 renormalized coupling @xmath16 and bare lattice coupling @xmath17 at one loop can be obtained by requiring equality of @xmath18 and @xmath19 up to one loop @xcite : s_ym(a)+_1(a)_=s_ym(a)+_1(a)_lat [ 2.5 ] expanding each side in powers of @xmath12 , the equality betwen the quadratic terms gives , in light of ( [ 2.4 ] ) , ) in the lattice setting up to terms which vanish for @xmath20 . ]
-_1(p)_=-_1(p)_lat [ 2.6 ] the one - loop contribution to the 2-point function is given by a sum of 3 terms : gauge field loop , ghost loop and fermion loop .
consequently , for @xmath21 flavors of massless fermions , @xmath22 and @xmath23 have the general forms _ 1(p)_&=&-_0(p^2/^2)+c_^g+n_fc_^f [ 2.7 ] + _ 1(p)_lat&=&-_0(a^2p^2)+c_lat^g+n_fc_lat^f [ 2.8 ] where @xmath24 is the mass scale in the @xmath0 scheme , @xmath4 is the lattice spacing , and _ 0=(n - n_f ) [ 2.9 ] the constant @xmath25 arises from the gauge and ghost loop contributions ; it depends on @xmath26 and the gauge fixing parameter , while @xmath27 arises from the fermion loop contribution . in the lattice case @xmath25 and @xmath27
also depend on whatever parameters are present in the lattice gauge and fermion actions .
( e.g. , for lattice wilson fermions @xmath28 depends on the wilson parameter . ) from ( [ 2.6 ] ) and ( [ 2.7])([2.8 ] ) we get = + _ 0(a^2
^ 2)+l_0 [ 2.10 ] where l_0&=&l_0^g+n_fl_0^f [ 2.11 ] + l_0^g = c_^g - c_lat^g & , & l_0^f = c_^f - c_lat^f [ 2.12 ] ( our notations @xmath29 , @xmath30 , @xmath31 follow the papers of panagopoulos and collaborators , e.g. ,
@xcite ) the relation bewteen @xmath16 and @xmath17 up to one loop now follows from ( [ 2.10 ] ) : g^2=g_0 ^ 2(1-g_0 ^ 2_0(a^2 ^ 2)+l_0+o(g_0 ^ 4 ) ) [ 2.13 ] also , from ( [ 2.10 ] ) or ( [ 2.13 ] ) the relation between the lattice and @xmath0 @xmath1-parameters is obtained @xcite : = e^l_0/2_0 [ 2.14 ] the focus of our attention in this paper is the fermionic contribution to @xmath29 , i.e. , @xmath31 in ( [ 2.12 ] ) for general lattice fermion formulation .
the continuum constant @xmath32 is well - known ( see , e.g. , @xcite ) : c_^f=- [ 2.15 ] therefore , to determine @xmath31 we need to determine the lattice constant @xmath28 for general lattice fermion formulation . to study @xmath28
it suffices to consider the @xmath33 case which we restrict to henceforth .
we proceed by expanding the lattice 2-point function @xmath34 in powers of ( the components of ) the external momentum @xmath35 . although @xmath34 itself is infrared finite , the terms in the expansion are individually infrared divergent . to deal with this
we introduce a fermion mass @xmath5 as regulator ; it renders the expansion terms finite . since gauge invariance is maintained when @xmath5 is introduced , the expansion of the one fermion loop contribution to @xmath34 up to 2nd order in @xmath35 results in an expression of the general form _ lat^(2)(p)_^ab_f,1-loop = ^ab(p^2_-p_p_)i(am)+r(p^2/m^2)[2.16 ] up to terms which vanish for lattice spacing @xmath20 .
here @xmath3 is a logarithmically divergent one loop lattice feynman integral given by i(am)=-___lat^(2)(p)_^11 _ f,1-loop|_p=0 [ 2.17 ] ( @xmath36 ) . by the structural result of @xcite it has the general form in the log term
is fixed by universality @xcite ; it is minus the fermionic term in @xmath37 with @xmath38 . ]
i(am)=(a^2m^2)+c_i [ 2.18 ] where the constant @xmath2 depends on the details of the lattice fermion formulation .
@xmath39 in ( [ 2.16 ] ) denotes the continuum limit of the remainder term after expanding @xmath40 to 2nd order in @xmath35 .
it is convergent by power - counting and therefore coincides @xcite with the corresponding ( known ) continuum term : r(p^2/m^2)&=&_0 ^ 1dxx(1-x)x(1-x)+1 + & = & ( p^2/m^2 ) + _ 0 ^ 1dxx(1-x)x(1-x)+ + & & ( p^2/m^2)+(m^2/p^2 ) [ 2.20 ] substituting ( [ 2.18 ] ) and ( [ 2.20 ] ) in ( [ 2.16 ] ) gives _
lat^(2)(p)_^ab_f,1-loop = ^ab(p^2_-p_p _ ) ( a^2p^2)+c_i+(m^2/p^2)[2.21 ] comparing this to the expression ( [ 2.4 ] ) for @xmath34 with @xmath23 given by ( [ 2.8 ] ) , we see that c_lat^f = c_i+(0 ) [ 2.22 ] in fact @xmath41 is precisely the @xmath0 constant @xmath42 : explicit evaluation of the integral in ( [ 2.20 ] ) at @xmath43 gives @xmath44 .
it follows from ( [ 2.12 ] ) that l_0^f =- c_i [ 2.23 ] thus the issue is to determine the constant @xmath2 appearing in the logarithmically divergent one - loop lattice integral @xmath3 in ( [ 2.17])([2.18 ] ) . in this paper
we are going to derive a compact general formula for @xmath3 for general lattice fermions which can then be used to determine @xmath2 .
in this section we derive an initial general formula for @xmath3 .
it is the starting point for deriving more explicit formulae in the subsequent sections .
let @xmath45 be a general lattice dirac operator which is translation - invariant and transforms covariantly under gauge transformations and rotations of the 4-dimensional euclidean hypercubic lattice .
the ansatz for the link variables is u_(x)=e^aa_(x+a ) [ 3.1 ] ( @xmath46=unit vector in the positive @xmath24-direction ) .
the fourier - transformed field @xmath47 is defined via a_(x+a)=_p_(p)e^ip(x+a ) .
[ 3.2 ] here and in the following @xmath48 .
the continuum gauge fields are @xmath49 where @xmath50 are generators for su(n ) normalized such that @xmath51 .
( we take the @xmath52 s to be anti - hermitian , absorbing into them the imaginary unit that multiplies @xmath12 in other notations . )
expanding the link variables in powers of @xmath12 gives an expansion of the lattice dirac operator , d=_n=0^d_n [ 3.3 ] translation - invariance implies that each @xmath53 can be expressed in momentum basis in the form d_n(k,k)&=&a^n-1_p_1, ,p_n(p_1+ +p_n+k - k ) d_n(ak|ap_1, ,ap_n)__1_n _ _
1(p_1)__n(p_n ) + & & [ 3.4 ] the continuum limit requirements on @xmath45 are d_0(ak)&&i_k_+m [ 3.4a ] + d_1(ak|ap ) _ & & _ [ 3.4b ] the mass @xmath5 enters via an additive term @xmath54 in @xmath55 ( we suppress the @xmath5-dependence in the notation ) . to derive a general formula for @xmath3 it is useful to exploit the fact , clear from ( [ 2.1 ] ) , that the one fermion loop contribution to @xmath8 is given by the fermion determinant : ( a)_f,1-loop =- d [ 3.5 ] where , in the present lattice setting , @xmath45 is coupled to the link variables ( [ 3.1 ] ) of the background gauge field @xmath12 . decomposing @xmath45 as d = d_0+d_int [ 3.6 ] where @xmath56 is the free field operator and @xmath57 the interaction part , we start from @xmath58 to obtain the expansion d - d_0&=&tr(1+d_0 ^ -1d_int ) + & = & _
l=1^tr(d_0 ^ -1d_int)^l[3.7 ] from this , using @xmath59 with @xmath53 given by ( [ 3.4 ] ) , an expansion in powers of @xmath12 is obtained .
it has the general form d - d_0=_n1_p_1, ,p_n-1 _ ^(n)(p_1, ,p_n-1)__1_n tr _ _
1(p_1)__n(p_n ) [ 3.8 ] ( @xmath60 ) .
the `` @xmath61-point function '' @xmath62 can be taylor - expanded in powers of ( the components of ) @xmath63 .
gauge invariance and mass - dimension considerations imply that the terms in the expansion can be combined to take the form ) to transform in the correct way under lattice gauge transformations the transformation of the continuum gauge field @xmath12 needs to be modified in the lattice setting @xcite .
however , one still finds that the mass - dimension zero function in ( [ 3.9 ] ) must be @xmath64 this can be inferred from the argument in t. reisz s proof of renormalizability of lattice qcd @xcite ; see also 5 of @xcite . ]
d - d_0=-i(am)s_ym(a)+_r1s_r(a ) [ 3.9 ] up to terms which vanish for @xmath20 , where each @xmath65 is a gauge invariant function with mass dimension @xmath66 .
the fact that the coefficient of @xmath67 is @xmath68 is inferred from ( [ 2.16 ] ) and ( [ 3.5 ] ) .
it is clear from ( [ 3.9 ] ) and ( [ 3.5 ] ) that @xmath3 can be evaluated from either the 2- , 3- or 4-point function in ( [ 3.8 ] ) . for the 4-point function
the relation is i(am)=-_^(4)(0,0,0 ) _ ,
[ 3.10 ] for any choice of @xmath24 and @xmath69 with @xmath70 .
usually @xmath3 is evaluated from the 2-point function via ( [ 2.17 ] ) , but we will see in the following that it is advantageous to use ( [ 3.10 ] ) instead . to evaluate it
we start by collecting the terms containing 4 powers of @xmath12 in the the expansion of @xmath71 in ( [ 3.7 ] ) . these are & & tr(d_0 ^ -1d_1)^4 + 3tr(d_0 ^ -1d_1)^2d_0 ^ -1d_2 + & & + ( 2tr(d_0 ^ -1d_1d_0 ^ -1d_3 ) + tr(d_0 ^ -1d_2)^2 ) + tr(d_0 ^ -1d_4 ) [ 3.11 ] inserting the expression ( [ 3.4 ] ) for each @xmath72 , and evaluating the traces in momentum basis , the @xmath73 in ( [ 3.7 ] ) is readily worked out .
the resulting expression for @xmath3 from ( [ 3.10 ] ) is as follows . from the functions @xmath74 in ( [ 3.4 ] ) we define d_n(k)__1_nd_n(k|0, ,0)__1_n [ 3.12 ] then i(am)=i_(1,1,1,1)(am)+i_(1,1,2)(am)+i_(2,2)(am)+i_(1,3)(am)+i_(0,4)(am ) [ 3.13 ] where & & i_(1,1,1,1)(am)=_-^tr d_0(k)^-1d_1(k)_d_0(k)^-1d_1(k)_d_0(k)^-1d_1(k)_d_0(k)^-1 d_1(k ) _ + & & [ 3.14 ] + & & i_(1,1,2)(am)=-_-^tr d_0(k)^-1d_1(k)_d_0(k)^-1d_1(k)_d_0(k)^-1d_2(k ) _ [ 3.15 ] + & & i_(2,2)(am)=_-^tr d_0(k)^-1d_2(k)_d_0(k)^-1d_2(k ) _ [ 3.16 ] + & & i_(1,3)(am)=_-^tr d_0(k)^-1d_1(k)_d_0(k)^-1d_3(k ) _ [ 3.17 ] + & & i_(0,4)(am)=-_-^trd_0(k)^-1d_4(k ) _ [ 3.18 ] for any choice of @xmath24 and @xmath69 with @xmath70 ( no sum over repeated indices ) .
the traces are over spinor indices alone .
each integral corresponds to an @xmath7 feynman diagram in fig .
[ vac1 ] in sect .
i and the notation reflects the structure of the corresponding diagram .
the above integrals could have been derived directly from the feynman diagrams , with fermion propagator and vertices read off from ( [ 3.4 ] ) , but it is easier ( and equivalent ) to use the fermion determinant expansion as we have done above .
the expression of @xmath3 as the sum of the integrals ( [ 3.14])([3.18 ] ) is our initial general formula .
it has a number of advantageous features which we note in the following . in light of ( [ 3.4a ] )
it is clear that the integral @xmath75 diverges logarithmically for @xmath76 while the other integrals ( [ 3.15])([3.18 ] ) are all finite in this limit .
the divergent integral necessarily @xcite has the general form i_(1,1,1,1)(am)=(a^2m^2)+_i [ 3.19 ] up to terms which vanish for @xmath20 .
then @xmath2 is given by c_i=_i+i_(1,1,2)(0)+i_(2,2)(0)+i_(1,3)(0)+i_(0,4)(0 ) [ 3.20 ] next we point out that @xmath75 , and hence @xmath77 , _ depend only on the free field lattice dirac operator_. this is because @xmath78 in ( [ 3.14 ] ) is determined by @xmath79 : for any lattice fermion formulation we have d_1(k)_=d_0(k ) . [ 3.21 ] this is derived in appendix a. therefore , a change of gauging of @xmath45 ( e.g. , by adding staples , clover term etc . ) affects only the finite integrals ( [ 3.15])([3.18 ] ) .
a further advantage becomes apparent when considering the description of @xmath45 in terms of lattice paths @xcite ( we will discuss the path description more explicitly in sect .
it is easy to see that the vertex function @xmath80 receives contributions only from lattice paths which contain a lattice link parallel to the @xmath81-axis , preceded at some point by a link parallel to the @xmath82-axis , preceded at some point by a link parallel to the @xmath83-axis , and so on . in particular , if the lattice paths describing @xmath45 are straight lines , as is the case for the wilson - dirac , naive and staggered operators , then @xmath84 vanishes unless the @xmath85 s are all the same .
it follows that the finite terms ( [ 3.15])([3.18 ] ) all vanish in this case ( since @xmath70 there ) .
thus , for such lattice dirac operators , @xmath3 is given entirely by @xmath75 .
if staples `` @xmath86 '' are attached to such an operator then ( [ 3.15])([3.17 ] ) are non - vanishing while ( [ 3.18 ] ) still vanishes . if a clover term ( @xmath87 closed paths around plaquettes ) is added then all the finite integrals ( [ 3.15])([3.18 ] ) are potentially non - vanishing .
however , as we will see in sect .
v , @xmath88 and @xmath89 both vanish in the clover case , so the terms ( [ 3.17])([3.18 ] ) also vanish there .
these properties make ( [ 3.13])([3.20 ] ) useful for evaluating @xmath3 and @xmath2 in practice for specific lattice fermion formulations , especially for improved formulations .
e.g. , when the improved @xmath45 differs from the original one by a more complicated choice of gauging the only new quantities that need to be evaluated are the finite integrals ( [ 3.15])([3.18 ] ) with @xmath90 in @xmath91 .
the quantities @xmath92 , @xmath88 and @xmath89 appearing in their integrands are generally easy to determine in practice as we will see in sect . v. besides these , the integrands only involve the free fermion @xmath91 and its derivative ( recall ( [ 3.21 ] ) ) . taking the initial formula for @xmath3 in this section as starting point , we go on to derive more explicit expressions in the following two sections .
for simplicity we use the notations @xmath93 and @xmath94 in the following .
we assume that _ 0d_0^d_0=d_0d_0^ [ 4.1 ] is a scalar , i.e. , trivial in spinor space .
note that this is a free field statement ; it holds for all lattice dirac operators of current interest ( naive , staggered , wilson , overlap , ... ) . recalling ( [ 3.21 ] ) : @xmath95 , and using the relations d_0
^ -1= , _ d_0
^ -1=(_d_0^ - ) , [ 4.2 ] evaluation of ( [ 3.14 ] ) leads to i_(1,1,1,1)(am)=_-^ [ 4.3 ] for any choice of @xmath96 with @xmath97 , where x_&= & -6d_0^_(_d_0_d_0^_d_0 ) + _ ^2(__0d_0^_d_0 ) -_^2(_0_d_0^_d_0 ) + & & + _ ( _ ^2_0d_0^_d_0 ) -2_^2_0_d_0^_d_0 [ 4.4 ] the details of the calculation are given in appendix b. to evaluate this expression further , we now assume that @xmath98 ( the free field momentum representation of the lattice dirac operator @xmath45 ) has the general form d_0=i__+[4.5 ] where @xmath99 and @xmath100 ( which includes the mass term ) are real scalar functions , and the gamma - matrices are hermitian , so that @xmath101 and @xmath102 .
this is the typical free field form of lattice dirac operators of interest in practice ; it covers the naive , staggered , wilson , and overlap operators and their improved versions . for simplicity
we make the further assumption that _ _ d_0=0 .
[ 4.6 ] this holds for wilson , naive and staggered fermions , also when the naik term is included , but does not hold for the overlap operator .
( the extention of the following to the general case @xmath103 is straightforward but tedious , and we defer it to a future article where the specific results for the overlap operator will be derived . ) substituting ( [ 4.5 ] ) into ( [ 4.3])([4.4 ] ) , a calculation using ( [ 4.6 ] ) and the other mentioned properties gives i_(1,1,1,1)(am)=_-^ ( ) [ 4.7 ] where y_&= & _
^2_^2_^2_^2 -12(__)^2(__)^2 + 2_^2 ^ 2_^2_^2 -24(_)^2(__)^2 + & & - 12_^2 _ _ _ _ -4(_)^2_^2 + 4 ^ 2_^2_^2 . [ 4.8 ] from this , using integration by parts ( in a similar way to the calculations in appendix b ) , a more compact expresion can be obtained : i_(1,1,1,1)(am)=-_-^ ( -(___0)^2 ) [ 4.8a ] ( recall @xmath102 ) .
although this expression is more compact , the preceding one ( [ 4.7])([4.8 ] ) seems more useful for evaluating @xmath75 in practice . in the remainder of this section
we show that these general formulae readily reproduce the previously known results for the specific cases of ( unimproved ) naive , staggered and wilson fermions .
( recall from sect .
iii that @xmath3 is given entirely by @xmath75 in these cases . ) for a naive fermion , @xmath104 , @xmath105 and ( [ 4.8 ] ) reduces to y_(k)=42k_2k_-12 ^ 2k_^2k _ [ 4.9 ] up to terms which are @xmath106 for @xmath20 . since a naive fermion decomposes into 4 degenerate staggered fermions @xcite , @xmath3 for a staggered fermion
is obtained by dividing the naive fermion result by @xmath107 .
the invariance of ( [ 4.9 ] ) under @xmath108 allows the integration domain in ( [ 4.7 ] ) to be restricted to @xmath109 ^ 4 $ ] at the expense of an overall factor 16 .
thus we find i(am)=n_t_-/2^/2 [ 4.10 ] up to terms which vanish for @xmath20 , where @xmath110 is the number of fermion `` tastes '' ( 16 for a naive fermion , 4 for a staggered fermion ) .
this is precisely the expression for @xmath3 derived previously in eq .
( 6.8 ) of @xcite where the contribution to the gluonic 2-point function from a naive / staggered fermion loop was evaluated .
( @xmath3 corresponds to @xmath111 in @xcite . ) after changing variables by @xmath112 the integral ( [ 4.10 ] ) can be evaluated by the method of ref.@xcite .
it expresses the integral in terms of certain basic lattice integrals that were evaluated numerically to high precision in @xcite .
we find the following result : i(am)=n_t((2am)^2+_e - f_0 ) + z_0 [ 4.12 ] where @xmath113 is the euler constant and @xmath114 , @xmath115 are numerical constants listed to high numerical accuracy in table 1 of @xcite .
thus the divergent part has the correct universal structure , and the constant @xmath2 ( @xmath116 ) that we are after is given by -c_i / n_t&=&(f_0-_e-(4 ) ) -z_0 + & = & 0.0026247621012431485 ... [ 4.13 ] in ref .
@xcite this constant was denoted @xmath117 and our value agrees with the one in eq .
( 6.16 ) of that paper .
we have obtained it to much higher numerical precision here though , thanks to the high accuracy to which @xmath118 and @xmath115 were evaluated in @xcite . in this case
@xmath104 , @xmath119 where @xmath120 is the wilson parameter , and ( [ 4.8 ] ) reduces to y_(k ) & = & 42k_2k_-12 ^ 2k_^2k _ + & & + r^2(8 ^ 2k_(2k_-3 ^ 2k_)-62k_k_k _ + 162k_k__^2(k_/2 ) ) + & & + r^4(-8 ^ 2k_k__^2(k_/2 ) + 16k_k_(_^2(k_/2))^2 ) + & & [ 4.14 ] up to terms which are @xmath106 for @xmath20 . substituting this into ( [ 4.7 ] ) we recover the previous result of kawai _ et al .
_ for the one fermion loop contribution in eq.(3.24)(3.25 ) of ref .
( dimensional regularization of the infrared divergence was used there , but the result is easily transformed into mass - regularized form ( cf . 5.1 of @xcite ) and then coincides with ours . )
the constant @xmath2 is related to the constant @xmath121 defined there by c_i=(_e-(4))+l [ 4.16 ] the integral @xmath3 can again be evaluated by the method of ref .
@xcite ; in fact this was done in that paper for @xmath122 , and the expression for @xmath121 in terms of the basic constants is given in eq.(6.8 ) of @xcite . from that , @xmath31 is obtained to high precision in the wilson fermion case : l_0^f =- c_i=0.00669599933173308 ... [ 4.17 ] the wilson fermion case was also considered independently by p. weisz in @xcite .
note that for wilson fermions @xmath123 , hence ( [ 4.8a ] ) simplifies to i(am)=-_-^ ( -(___0)^2 ) [ 4.18 ] thus reproducing weisz s expression in eq .
( 13 ) of @xcite .
the constant @xmath31 , denoted as @xmath124 there , was evaluated numerically but to low precision
a systematic error of @xmath125 was estimated , and this is indeed the case when comparing the value for @xmath124 in eq.(22 ) of @xcite with the high precision value in ( [ 4.17 ] ) above .
in this section we derive general formulae for the finite integrals ( [ 3.15])([3.18 ] ) and describe how these can be straightforwardly evaluated , using staple and clover terms as illustrations .
_ evaluation of @xmath126 . _
starting from ( [ 3.15 ] ) , calculations of the same type as in appendix b lead to i_(1,1,2)(am)=-_-^ [ 5.1 ] ( for any choice of @xmath96 with @xmath70 ) with the notations @xmath127 , @xmath128 and @xmath129 .
specializing as before to the case where @xmath98 has the general form @xmath130 and satisfies @xmath131 for @xmath70 , and _ assuming _ that the gamma - matrix structure of @xmath132 has the general form d_2(k)_=__e_(k ) ( ) [ 5.2 ] ( [ 5.1 ] ) leads to i_(1,1,2)(am)=-_-^ [ 5.3 ] ( there are no terms involving derivatives of @xmath133 since these all cancel out . ) _ evaluation of @xmath134 . _ from ( [ 3.16 ] ) we obtain i_(2,2)(am)=_-^ [ 5.4 ] specializing as before , this leads to i_(2,2)(am)=-2_-^ [ 5.5 ] _ evaluation of @xmath135 . _ from ( [ 3.17 ] ) we find i_(1,3)(am)=_-^ [ 5.6 ] with the notation @xmath136 .
specializing as before , and assuming that @xmath137 has the general form d_3(k)_=_e_-if _ [ 5.7 ] ( [ 5.6 ] ) leads to i_(1,3)(am)=4_-^ [ 5.8 ] _ evaluation of @xmath138 . _ from ( [ 3.18 ] ) we find i_(0,4)(am)=-_-^ [ 5.9 ] with the notation @xmath139 .
a specialized formula can be worked out as in the previous cases ( we omit the details ) .
note that terms in @xmath140 involving a product of two or more gamma matrices give vanishing contribution to the trace in ( [ 5.9 ] ) . to evaluate the integrals in practice one needs to determine @xmath92 , @xmath88 and @xmath141 ( @xmath70 ) for the lattice dirac operator @xmath45 .
this can be done straightforwardly from the description of @xmath45 in terms of lattice paths , as we now describe . in the path description , @xmath45 is expressed as d(x)=_c__u_x , x+a_ ( x+a_ ) [ 5.10 ] where the sum is over a collection of translation - equivalence classes @xmath142 of oriented lattice paths .
each has an associated numerical constant @xmath143 and element @xmath144 of the clifford algebra generated by the gamma - matrices . associated with each equivalence class @xmath142 is a vector @xmath145 with integer components : it is the difference in lattice units betwen the start- and end - points .
@xmath146 denotes the product of the link variables along the representative path for @xmath142 starting at @xmath147 and ending at @xmath148 .
the clifford algebra - valued functions @xmath149 in the expansion ( [ 3.3])([3.4 ] ) of @xmath45 are found by expanding the link variable products @xmath146 in ( [ 5.10 ] ) in powers of the continuum gauge field @xmath12 . in the present case
we only need to determine these functions at vanishing `` external momenta '' , i.e. , @xmath150 .
therefore we can take the continuum gauge fields @xmath151 to be constants
. then @xmath146 is independent of @xmath148 , and its expansion in powers of the continuum gauge field has the form u_=_n u__1_n^a__1a__n [ 5.11 ] the functions @xmath84 are then given by d_n(k)__1_n=_d_n^(k)__1_n [ 5.12 ] where d_n^(k)__1_n = c__u__1_n^ e^i_k [ 5.13 ] thus , given the description ( [ 5.10 ] ) of @xmath45 in terms of lattice paths , the problem of determining @xmath84 is reduced to determining the coefficients @xmath152 in the expansion ( [ 5.11 ] ) of the link variable product @xmath153 with constant @xmath12 , which is generally straightforward in practice . in the present case things
are further simplified since we only need to know @xmath84 when @xmath154 is @xmath155 , @xmath156 , and @xmath157 .
since @xmath70 we can replace each link variable @xmath158 appearing in the product @xmath153 by u_^1 \ { ll 1a _ & + 1 & + .
[ 5.14 ] to illustrate this , consider the case of a `` @xmath159 staple '' @xmath160
. the relevant parts of the expansion are found by u_u_u_^-1&&(1+a_)(1+a_)(1-a _ ) + & & = -a_-a_a_a_+ [ 5.15 ] where `` other '' refers to terms which are not proportional to @xmath161 , @xmath162 or @xmath163 and therefore do not contribute to @xmath92 , @xmath164 or @xmath141 .
thus the relevant contributions from the staple are d_2^(k)_=-c__e^ik_,d_3^(k)_=-c__e^ik_,d_4^(k)_=0 [ 5.16 ] where @xmath165 are whatever factors that accompany the staple @xmath153 in the expression ( [ 5.10 ] ) for the lattice dirac operator .
as another example we now consider the sheikholeslami - wohlert clover term for wilson clover fermions @xcite : c = c_sw__p _ [ 5.17 ] where @xmath166 is a tunable constant , @xmath167 , and @xmath168 is a sum of products of link variables around oriented plaquettes in the @xmath169-plane starting and ending at @xmath148 .
the relevant contributions in this case come from the @xmath170 parts : c_sw(_p_+_p _ ) = -c_swp _ [ 5.18 ] for constant gauge fields , @xmath171 is explicitly given by p_= l u_u_u_^-1u_^-1+u_u_^-1u_^-1u _
+ + u_^-1u_^-1u_u_+u_^-1u_u_u_^-1 + - ( ) + [ 5.19 ] expanding this as described above , the relevant expansion coefficients are found to be @xmath172 , @xmath173 , and @xmath174 .
since the plaquette paths are closed we have @xmath175 in all cases . using this
, it follows from ( [ 5.12])([5.13 ] ) that d_2(k)_=-c_sw , d_3(k)_=0,d_4(k)_=0 [ 5.20 ] independent of @xmath176 .
note that , as discussed in sect .
iii , there are no contributions from the wilson - dirac operator since in its path description the paths are all straight lines .
the finite integral contributions to @xmath3 can now be determined in the case of wilson clover fermions . by (
[ 5.20 ] ) and the previous formulae , @xmath177 .
the other integrals are determined by substituting @xmath178 , @xmath179 and @xmath180 into ( [ 5.3 ] ) and ( [ 5.5 ] ) .
taking the wilson parameter to be @xmath122 and evaluating the integrals by the method of ref .
@xcite we find , in the @xmath20 limit , i_(1,1,2)(0)/c_sw&=&(1,0)-(2,-1 ) + & = & 0.005046714024535753066 ... [ 5.21 ] and i_(2,2)(0)/c_sw^2&=&-(2,-2 ) + & = & -0.02984346719542684815 ... [ 5.22 ] where @xmath181 , @xmath182 and @xmath183 are certain basic convergent one loop lattice integrals defined in 4 of ref .
@xcite , whose numerical values are given to high precision in table 2 of that paper .
recalling ( [ 3.20 ] ) , collecting the results for wilson clover fermions with @xmath122 we have l_0^f =- c_i =- c_sw - c_sw^2 [ 5.23 ] ( ( [ 4.17 ] ) means the numerical constant given in eq .
( [ 4.17 ] ) , etc . )
this agrees with the previous literature but gives the numerical constants to much greater precision .
the previously most precise values were those in eq .
( 14 ) of ref .
@xcite where ( [ 5.21 ] ) and ( [ 5.22 ] ) were given to 11 decimal places .
we have obtained them here to 20 decimal places , thanks to the high precision with which the basic integrals in ref .
@xcite were evaluated . for earlier results for these quantities ,
see @xcite and the references therein .
the focus in this paper has been on deriving the general integral formulae for @xmath3 , confirming their correctness by checking that they reproduce the known results in the cases of staggered , wilson and clover fermions , and in the process developing general techniques for evaluating the formulae . in doing this
we were able to express @xmath3 in those cases in terms of basic one loop lattice integrals that have already been evaluated to high precision in @xcite .
this had already been done in @xcite in the wilson fermion case , but the results for the staggered and clover cases ( ( [ 4.13 ] ) and ( [ 5.21])([5.22 ] ) , respectively ) are presented here for the first time .
the further applications of the general formulae are left for future work , and in the following we discuss some possibilities for this .
obvious targets for future applications are the various improved versions of staggered fermions .
these formulations involve `` smearing '' of the link variables to reduce flavor symmetry violations ; specifically there is the `` fat-7 '' link @xcite and hyp link @xcite .
since these differ from unimproved staggered fermions only by a choice of gauging , evaluating @xmath3 should be straighforward : expand the relevant products of link variables in powers of the constant continuum gauge field to determine @xmath92 , @xmath88 and @xmath89 as described in sect .
v ; then , from the formulae in sect .
v the convergent integrals @xmath184 , @xmath185 , @xmath186 and @xmath187 can be explicitly evaluated by the method of ref .
@xcite ; this will express the integrals in terms of basic lattice integrals that have already been calculated to high precision . adding these to the already known @xmath75 ( @xmath188 for unimproved staggered fermions ) then gives @xmath3 in these improved cases . of more interest , however , is the case of @xmath189 improved `` asqtad '' staggered fermions @xcite that are currently used by the milc collaboration to generate the ensembles used in high precision lattice simulations @xcite .
besides smearing of link variables , this formulation also contains the naik term @xcite which modifies the free field staggered dirac operator .
therefore , @xmath190 is not the same as for unimproved staggered fermions in this case , and the method and results of ref .
@xcite do not apply .
@xmath3 may still be readily determined in this case from the general formulae and techniques of this paper , but it will be necessary to numerically evaluate the one loop lattice integrals that arise .
( the approach of ref .
@xcite could be used for this . )
the relation between the @xmath0 coupling and bare lattice coupling has already been determined to 2 loops for asqtad staggered fermions via symbolic computer program @xcite .
determining the fermionic contribution to the one loop coefficient independently via the ( semi-)analytic approach of the present paper would provide a useful check on the computer program .
another future application is to overlap fermions .
the fermionic contribution to the one loop relation between the @xmath0 coupling and bare lattice coupling in this case was calculated via symbolic computer program in @xcite , and the two loop relation was subsequently calculated in @xcite .
application of the results of the present paper will allow the one fermion loop contribution to be obtained from numerical evaluation of one loop lattice integrals without the need for symbolic computer programs . reproducing the result of @xcite in this way will be a useful check on the computer program , which the two loop result @xcite also relies on .
it will also allow the one fermion loop contribution to be calculated more easily and to higher precision for any value of the overlap parameter that one wishes to consider .
i emphasize that , in the approach of the present paper , the technical problem of calculating the one fermion loop contribution for overlap fermions is greatly reduced compared to the usual approach followed in @xcite .
the simplification comes about because the determination of the quantities @xmath92 , @xmath88 and @xmath89 in the expansion of the overlap dirac operator can be done with constant continuum gauge fields .
expansion of the overlap dirac operator in powers of constant continuum gauge fields is relatively straightforward , and has already been successfully used @xcite to reproduce results obtained via symbolic computer program in another context @xcite .
having found useful general formulae for the fermionic contribution to the coefficient in the one loop relation between the @xmath0 and bare lattice couplings , a natural question is whether a similar approach is possible for the contribution from the gauge and ghost loops . in fact
this seems quite possible . in the background field
approach the gauge field in the action is @xmath11 , and it is the part quadratic in the quantum fluctuation fields @xmath13 that determines the one loop contribution to the effective action . coming from the functional integral of a quadratic term , it is clear that this can be expressed as a functional determinant , both in the continuum and lattice settings .
there is also the faddeev - popov determinant through which the ghosts arise .
one can then expand the determinants in powers of @xmath12 as done for the fermion determinant in this paper .
the constant to be determined in this case , @xmath191 in ( [ 2.8 ] ) , can again be found from the quartic term in the expansion ( via ward indentities ) . in the fermionic case considered in the present paper ,
the possibility to introduce an infrared regulator mass was crucial for deriving the general formulae .
this can also be done in the gluonic
ghost case : note that @xmath13 transforms under gauge transformations by @xmath192 so a regulator mass term @xmath193 may be introduced without breaking gauge invariance .
this is currently being pursued , and i hope to be able to present general formulae for the gluonic ghost contribution to the one loop coefficient in future work .
the hope is that this may allow allow a ( semi-)analytic calculation of the one loop contribution in the case of improved lattice gauge actions , which has recently been evaluated via symbolic computer program in @xcite . finally , the results of this paper are relevant for a previous proposal for constructing the gauge field action on the lattice from the lattice fermion determinant @xcite . set @xmath194 and regard @xmath195 as a function of @xmath196 and @xmath4 .
expanding @xmath62 in ( [ 3.8 ] ) in powers of the momenta , without taking the @xmath20 limit , leads on dimensional grounds to the following variant of ( [ 3.9 ] ) : d()-d_0()=-i()s_ym(a)+_r1a^r_r(a , ) [ 6.1 ] where @xmath197 , a function of @xmath12 and @xmath196 , has mass - dimension @xmath120 . now taking @xmath20 we obtain , taking @xmath20 with @xmath196 fixed amounts to taking a simultaneous continuum and large mass limit
the fact that the large mass limit of the lattice fermion determinant effectively gives a lattice gauge action was discussed before in @xcite . ] _
a0 ( d()-d_0())=-i()s_ym(a ) [ 6.2 ] ( strictly speaking this depends on the sum in ( [ 6.1 ] ) being convergent , which requires the continuum gauge field @xmath12 to be sufficiently weak . )
thus the coefficient of the ym gauge action obtained from the lattice fermion determinant is seen to be @xmath198 , which can be evaluated from the general formulae in this paper after replacing @xmath54 by @xmath196 .
and the coefficient @xmath199 in @xcite is @xmath200 . ] in the @xmath201 limit we have i()(^2)+c_i [ 6.3 ] although calculation of the constant @xmath2 was the main focus in this paper , the general formulae can just as well be used to determine @xmath202 for general values of @xmath196 .
the integrals will need to be evaluated numerically for the specific values of @xmath196 that one considers though .
i am very grateful to prof .
peter weisz for his constructive suggestions on the previous ( and completely different ) version of this paper . at s.n.u .
the author is supported by the bk21 program . at f.i.u .
the author was supported under nsf grant phy-0400402 .
some of this work was done during the kitp program `` modern challenges for lattice field theory '' where the author was supported under nsf grant phy99 - 0794 .
we derive this from the path description ( [ 5.10])([5.13 ] ) of the lattice dirac operator . from ( [ 5.13 ] ) we have d_0^(k)&=&c__e^i_k [ a1 ] + d_1^(k)_&=&c__u_^e^i_k [ a2 ] thus to show the relation ( [ 3.21 ] ) : @xmath203 it suffices to show u_^=(_ ) _ [ a3 ] from the definition ( [ 5.11 ] ) it is easy to see that @xmath204 counts the number of links of the path @xmath142 that lie along the @xmath24-direction , counted with sign @xmath205 depending on whether they are oriented in the positive or negative @xmath24-direction . but this is precisely the @xmath24-coordinate of the difference ( in lattice units ) between the start- and end points of ( a representative path for ) @xmath142 , i.e. , the @xmath24-coordinate of @xmath145 .
thus we have found that ( [ a3 ] ) holds .
the general formula ( [ 4.3])([4.4 ] ) is obtained by evaluating the integrand in the initial expression ( [ 3.14 ] ) as follows . for notational simplicity
we omit the `` @xmath206 '' , write @xmath207 , @xmath208 for @xmath98 , @xmath209 , and use `` @xmath87 '' to denote equality up to terms which vanish upon taking the trace or terms which are total derivatives and therefore give vanishing contribution to the integral . & & d^-1dd^-1dd^-1dd^-1d + & & = d^-1dd^-1d + & & = ( d^- ) d(d^-)d + & & d^dd^d + ( ) d^dd^d -()d^dd^d + & & -d^(dd^d ) + ( ^2d^dd^d + ( d^dd^d ) ) + & & [ b.1 ] the second term here is re - expressed using d^dd^d = ( -d^d)d^d = d^d - d^d [ b.2 ] to get ( ) ^2d^d -^2d^d & & ( ( ^2d^d ) -^2d^d ) + & & [ b.3 ] the third term in ( [ b.1 ] ) is re - expressed as ( ) ( d^dd^d ) ^2(d^dd^d ) [ b.4 ] the resulting expression for the integrand , given by the first term in ( [ b.1 ] ) plus ( [ b.3 ] ) plus ( [ b.4 ] ) , leads to the claimed formula ( [ 4.3])([4.4 ] ) . g.p .
lepage and p.b .
mackenzie , phys .
d 48 ( 1993 ) 2250 [ hep - lat/9209022 ] h.d .
trottier , nucl .
suppl . 129 ( 2004 ) 142 [ hep - lat/0310044 ] m. lscher and p. weisz , nucl .
b 452 ( 1995 ) 234 [ hep - lat/9505011 ] a. bode , p. weisz and u. wolff , nucl .
b 576 ( 2000 ) 517 ; erratum - ibid .
b 600 ( 2001 ) 453 [ hep - lat/9911018 ] g. martinelli and c.t .
sachrajda , nucl .
b559 ( 1999 ) 429 [ hep - lat/9812001 ] q. mason _
[ hpqcd collaboration ] , phys .
d 73 ( 2006 ) 114501 [ hep - ph/0511160 ] a. hasenfratz and p. hasenfratz , phys .
b 93 ( 1980 ) 165 .
r. dashen and d. gross , phys .
d 23 ( 1981 ) 2340 .
a. hasenfratz and p. hasenfratz , nucl .
b 193 ( 1981 ) 210 h. kawai , r. nakayama and k. seo , nucl .
b 189 ( 1981 ) 40 .
p. weisz , phys .
b 100 ( 1981 ) 331 .
sharatchandra , h.j .
thun and p. weisz , nucl .
b 192 , 205 ( 1981 ) . c. alexandrou , h. panagopoulos and e. vicari , nucl .
b 571 ( 2000 ) 257 [ hep - lat/9909158 ] .
a. skouroupathis and h. panagopoulos , phys .
d 76 ( 2007 ) 114514 [ arxiv:0709.3239 ] b. alls , a. feo and h. panagopoulos , nucl .
b 491 ( 1997 ) 498 [ hep - lat/9609025 ] c. christou , a. feo , h. panagopoulos and e. vicari , nucl .
b525 ( 1998 ) 387 ; erratum - ibid .
b 608 ( 2001 ) 479 [ hep - lat/9801007 ] a. bode and h. panagopoulos , nucl .
b 625 ( 2002 ) 198 [ hep - lat/0110211 ] m. constantinou and h. panagopoulos , phys
. rev .
d 76 ( 2007 ) 114504 [ arxiv:0709.4368 ] m. constantinou and h. panagopoulos , phys .
d 77 ( 2008 ) 057503 [ arxiv:0711.4665 ] s. capitani , phys .
rept . 382 ( 2003 ) 113 [ hep - lat/0211036 ] .
b. sheikholeslami and r. wohlert , nucl .
b 259 ( 1985 ) 572 .
s. naik , nucl .
b 316 ( 1989 ) 238 g. burgio , s. caracciolo and a. pelissetto , nucl .
b 478 ( 1996 ) 687 [ hep - lat/9607010 ] .
lepage , phys .
d 59 ( 1999 ) 074502 [ hep - lat/9809157 ] h. neuberger , phys .
b 417 ( 1998 ) 141 [ hep - lat/9707022 ] ; i. horvath , hep - lat/0607031 d.h .
adams and w. lee , phys .
d 77 ( 2008 ) 045010 [ arxiv:0709.0781 ] t. reisz , comm .
phys . 116 ( 1988 ) 81 .
t. reisz , nucl .
b 318 ( 1989 ) 417 .
p. hasenfratz _ et al .
_ , int . j. mod
. phys .
c 12 ( 2001 ) 691 [ hep - lat/0003013 ] c. gattringer , phys .
d 63 ( 2001 ) 114501 [ hep - lat/0003005 ] n. kawamoto and j. smit , nucl .
b 192 ( 1981 ) 100 k. orginos and d. toussaint , phys .
d 59 ( 1999 ) 014501 [ hep - lat/9805009 ] a. hasenfratz and f. knechtli , phys .
d 64 ( 2001 ) 034504 [ hep - lat/0103029 ] c.t.h .
davies _ et al . _
[ hpqcd , ukqcd , milc and fermilab collaborations ] , phys . rev .
lett . 92 ( 2004 ) 022001 [ hep - lat/0304004 ] .
t. becher and k. melnikov , phys .
d 66 ( 2002 ) 074508 [ hep - ph/0207201 ] d.h .
adams , `` lattice gauge action from overlap fermions '' ( private notes , to appear as an article ) . | a compact general integral formula is derived from which the fermionic contribution to the one - loop coefficient in the perturbative expansion of the @xmath0 coupling in powers of the bare lattice coupling can be extracted .
it is seen to reproduce the known results for unimproved naive , staggered and wilson fermions , and has advantageous features which facilitate the evaluation in the case of improved lattice fermion formulations .
this is illustrated in the case of wilson clover fermions , and an expression in terms of known lattice integrals is obtained in this case which gives the coefficient to much greater numerical accuracy than in the previous literature . |
a fundamental problem of broad theoretical and practical interest is to characterize the maximum correlation between the outputs of a pair of functions of random sequences .
consider the two distributed agents shown in figure [ fig : agents ] .
a pair of correlated discrete memoryless sources ( dms ) are fed to the two agents .
these agents are to each make a binary decision .
the goal of the problem is to maximize the correlation between the outputs of these agents subject to specific constraints on the decision functions .
the study of this setup has had impact on a variety of disciplines , for instance , by taking the agents to be two encoders in the distributed source coding problem @xcite , or two transmitters in the interference channel problem @xcite , or alice and bob in a secret key - generation problem @xcite , or two agents in a distributed control problem @xcite .
a special case of the problem is the study of common - information ( ci ) generated by the two agents . as an example , consider two encoders in a slepian - wolf ( sw ) setup .
let @xmath0 , and @xmath1 be independent , non - constant binary random variables .
then , an encoder observing the dms @xmath2 , and an encoder observing @xmath3 agree on the value of @xmath1 with probability one .
the random variable @xmath1 is called the ci observed by the two encoders .
these encoders require a sum - rate equal to @xmath4 to transmit the source to the decoder .
this gives a reduction in rate equal to the entropy of @xmath1 , compared to the transmission of the sources over independent point - to - point channels .
the gain in performance is directly related to the entropy of the ci .
so , it is desirable to maximize the entropy of the ci between the encoders .
in @xcite , the authors investigated multi - letterization as a method for increasing the ci .
they showed that multi - letterization does not lead to an increase in the ci .
more precisely , they prove the following statement : _ let @xmath5 and @xmath6 be two sequences of dmss .
let @xmath7 and @xmath8 be two sequences of functions which converge to one another in probability .
then , the normalized entropies @xmath9 , and @xmath10 are less than or equal to the entropy of the ci between @xmath5 and @xmath6 for large @xmath11 . _
a stronger version of the result was proved by witsenhausen @xcite , where maximum correlation between the outputs is upper - bounded subject to the following restrictions on the decision functions : + _ 1 ) the entropy of the binary output is fixed .
+ 2 ) the agents cooperate with each other .
_ it was shown that maximum correlation is achieved if both users output a single element of the string without further processing ( e.g. each user outputs the first element of its corresponding string ) .
this was used to conclude that common - information can not be induced by multi - letterization . while , the result was used extensively in a variety of areas such as information theory , security , and control @xcite , in many problems , there are additional constraints on the set of admissible decision functions . for example
, one can consider constraints on the ` effective length ' of the decision functions .
this is a valid assumption , for instance , in the case of communication systems , the users have lower - bounds on their effective lengths due to the rate - distortion requirements in the problem @xcite . in this paper ,
the problem under these additional constraints is considered . a new upper - bound on the correlation between the outputs of arbitrary pairs of boolean functions
is derived .
the bound is presented as a function of the dependency spectrum of the boolean functions .
this is done in several steps .
first , the effective length of an additive boolean function is defined .
then , we use a method similar to @xcite , and map the boolean functions to the set of real - valued functions .
using tools in real analysis , we find an additive decomposition of these functions .
the decomposition components have well - defined effective lengths . using the decomposition
we find the dependency spectrum of the boolean function .
the dependency spectrum is a generalization of the effective length and is defined for non - additive boolean functions .
lastly , we use the dependency spectrum to derive the new upper - bound .
the rest of the paper is organized as follows : section [ sec : not ] presents the notation used in the paper .
section [ sec : eff ] develops useful mathematical machinery to analyze boolean function .
section [ sec : corr ] contains the main result of the paper .
finally , section [ sec : con ] concludes the paper .
in this section , we introduce the notation used in this paper .
we represent random variables by capital letters such as @xmath12 .
sets are denoted by calligraphic letters such as @xmath13 .
particularly , the set of natural numbers and real numbers are shown by @xmath14 , and @xmath15 , respectively . for random variables ,
the @xmath11-length vector @xmath16 is denoted by @xmath17 . the binary string @xmath18 is written as @xmath19 .
the vector of random variables @xmath20 , j_i\neq j_k$ ] , is denoted by @xmath21 , where @xmath22 $ ] .
for example , take @xmath23 , the vector @xmath24 is denoted by @xmath25 , and the vector @xmath26 by @xmath27 . for two binary strings @xmath28 , we write @xmath29 if and only if @xmath30 $ ] . for a binary string
@xmath19 we define @xmath31 , where @xmath32 denotes the hamming weight .
lastly , the vector @xmath33 is the element - wise complement of @xmath19 .
in this section , we study the correlation between the output of a boolean function with subsets of the input .
particularly , we are interested in the answers to questions such as ` how strongly does the first element @xmath34 affect the output of @xmath35 ? ' ` is this effect amplified when we take @xmath36 into account as well ? ' ` is there a subset of random variables that ( almost ) determines the value of the output ? ' .
we formulate these questions in mathematical terms , and find a characterization of the dependency spectrum of a boolean function .
the dependency spectrum is a vector which captures the correlation between different subsets of the input elements with each element of the output . as an intermediate step ,
we define the effective length of an additive boolean function below : for a boolean function @xmath37 defined by @xmath38 $ ] , where the addition operator is the binary addition , the effective length is defined as the cardinality of the set @xmath39 . for a general boolean function ( e.g. non - additive ) , we find a decomposition of @xmath40 into a set of functions @xmath41 whose effective length is well - defined .
first , we provide a mapping from the set of boolean functions to the set of real functions .
this allows us to use the tools available in real analysis to analyze these functions .
fix a discrete memoryless source @xmath5 , and a boolean function defined by @xmath42 .
let @xmath43 .
the real - valued function corresponding to @xmath44 is represented by @xmath45 , and is defined as follows : @xmath46 note that @xmath45 has zero mean and variance @xmath47 .
[ rem : exp_0 ] the random variable @xmath48 has finite variance on the probability space @xmath49 .
the set of all such functions is denoted by @xmath50 .
more precisely , we define @xmath51 as the separable hilbert space of all measurable functions @xmath52 . since x is a dms , the isomorphy relation @xmath53 holds @xcite , where @xmath54 indicates the tensor product .
the hilbert space @xmath55 is the space of all measurable functions @xmath56 .
the space is spanned by the two linearly independent functions @xmath57 and @xmath58 , where @xmath59 .
we conclude that the space is two - dimensional .
[ ex : ex1 ] the tensor operation in @xmath50 is real multiplication ( i.e. @xmath60 ) .
let @xmath61\}$ ] be a basis for @xmath55 , then a basis for @xmath50 would be the set of all the real multiplications of these basis elements : @xmath62}f_{i_j}(x_j ) , i_j\in [ 1,d]\}$ ] .
example [ ex : ex1 ] gives a decomposition of the space @xmath55 .
next , we introduce another decomposition of @xmath55 which turns out to be very useful .
let @xmath63 be the subset of all measurable functions of @xmath5 which have 0 mean , and let @xmath64 be the set of constant real functions of @xmath5 .
we argue that @xmath65 gives a decomposition of @xmath55 .
@xmath63 and @xmath64 are linear subspaces of @xmath55 .
@xmath63 is the null space of the linear functional which takes an arbitrary function @xmath66 to its expected value @xmath67 .
the null space of any non - zero linear functional is a hyper - space in @xmath55 .
so , @xmath63 is a one - dimensional subspace of @xmath55 . from remark [ rem : exp_0 ] , @xmath68 .
we conclude that any element of @xmath63 can be written as @xmath69 .
@xmath64 is also one dimensional .
it is spanned by the function @xmath70 .
consider an arbitrary element @xmath66 .
one can write @xmath71 where @xmath72 , and @xmath73 . [
ex : one_dim ] replacing @xmath55 with @xmath74 in , we have : @xmath75 where @xmath76 and , in ( a ) , we have used the distributive property of tensor products over direct sums .
equation , can be interpreted as follows : for any @xmath77 , we can find a decomposition @xmath78 , where @xmath79 .
@xmath80 can be viewed as the component of @xmath45 which is only a function of @xmath81 . in this sense ,
the collection @xmath82}i_j = k\}$ ] , is the set of components of @xmath45 whose effective length is @xmath83 .
[ rem : dec ] in order clarify the notation , we provide the following example : let @xmath5 be a binary symmetric source , and let @xmath84 be the binary ` and ' function .
the corresponding real function is : @xmath85 lagrange interpolation gives @xmath86 .
the decomposition is given by : @xmath87 the variances of these functions are given below : @xmath88 as we shall see in the next section , these variances play a major role in determining the correlation preserving properties of @xmath45 .
the vector whose elements include these variances is called the dependency spectrum of @xmath44 . in the perspective of the effective length , the function @xmath45 has @xmath89 of its variance distributed between @xmath90 , and @xmath91 which have effective length one , and @xmath92 of the variance is on @xmath93 which is has effective length two .
similar to the above examples , for arbitrary @xmath77 , we find a decomposition @xmath78 , where @xmath79 .
we characterize @xmath80 in terms of products of the basis elements of @xmath94}\mathcal{g}_{i_j}$ ] using the following result in linear algebra : let @xmath95 $ ] be vector spaces over a field @xmath96 .
also , let @xmath97\}$ ] be the basis for @xmath98 where @xmath99 is the dimension of @xmath98 .
then , any element @xmath100}\mathcal{h}_i$ ] can be written as @xmath101}\sum_{j_2\in [ 1,d_2]}\cdots \sum_{j_n\in [ 1,d_n ] } c_{j^n } v_{j_1}\otimes v_{j_2}\cdots \otimes v_{j_n}$ ] .
[ lem : tensor_dec ] since @xmath102 s , @xmath103 $ ] take values from the set @xmath104 , they are all one - dimensional . for the binary source @xmath5 with @xmath105 , define @xmath106 as : @xmath107 then , the single element set @xmath108 is a basis for @xmath63 . also , the function @xmath109 spans @xmath64 .
so , using lemma [ lem : tensor_dec ] , @xmath110 .
we are interested in the variance of @xmath80 s . in the next proposition , we show that the @xmath80 s are uncorrelated and we derive an expression for the variance of @xmath80 .
[ pr : partfun ] define @xmath111 as the variance of @xmath80 .
the following hold : + 1 ) @xmath112 , in other words @xmath80 s are uncorrelated .
@xmath113 \1 ) follows by direct calculation .
2 ) holds from the independence of @xmath114 s .
next , we find the characterization for @xmath80 .
@xmath115 gives the unique orthogonal decomposition of @xmath45 into the hilbert spaces @xmath116 .
[ lem : unique ] please refer to the appendix .
the following example clarifies the notation used in lemma [ lem : unique ] .
consider the case where @xmath117 .
we have the following decomposition of @xmath118 : @xmath119 let @xmath120 be an arbitrary function in @xmath118 .
the unique decomposition of @xmath45 in the form given in is as follows : @xmath121 it is straightforward to show that each of the @xmath122 s , @xmath123 , belong to their corresponding subspaces . for instance
, @xmath124 is constant in @xmath34 , and is a @xmath125 mean function of @xmath36 ( i.e. @xmath126 ) , so @xmath127 .
the following proposition describes some of the properties of @xmath80 which were derived in the proof of lemma [ lem : unique ] : [ prop : belong2 ] the following hold : + 1 ) @xmath128=0 .
+ 2 ) @xmath129 , we have @xmath130 .
+ 3 ) @xmath131 , for @xmath132 .
+ 4 ) @xmath133
lastly , we derive an expression for @xmath111 : for arbitrary @xmath37 , let @xmath45 be the corresponding real function , and let @xmath78 be the decomposition in the form of equation .
the variance of each component in the decomposition is given by the following recursive formula @xmath134 , where @xmath135 .
[ lem : power ] @xmath136 where ( a ) follows from 1 ) in proposition [ prop : belong2 ] , b ) follows from the decomposition in equation , ( c ) uses linearity of expectation , ( d ) uses 4 ) in proposition [ prop : belong2 ] , ( e ) holds from 2 ) in [ prop : belong2 ] , and in ( f ) and ( g ) we have used 1 ) in proposition [ prop : belong2 ] . for an arbitrary @xmath37 with corresponding real function @xmath45 , and decomposition @xmath137 .
let the variance of @xmath45 be denoted by @xmath138 .
then , @xmath139 .
[ cor : power ] the corollary is a special case of lemma [ lem : power ] , where we have taken @xmath19 to be the all ones vector .
the following provides a definition of the dependency spectrum of a boolean function : for a boolean function @xmath44 , the vector of variances @xmath140 is called the dependency spectrum of @xmath44 . in the next section , we will use the dependency spectrum to upper - bound the maximum correlation between the outputs of two arbitrary boolean functions .
we proceed with presenting the main result of this paper .
let @xmath141 be a pair of dms s . consider two arbitrary boolean functions @xmath142 and @xmath143 .
let @xmath144 , @xmath145 .
let @xmath146 , and @xmath147 give the decomposition of these functions as defined in the previous section .
the following theorem provides an upper - bound on the probability of equality of @xmath35 and @xmath148 .
let @xmath149 , the following bound holds : @xmath150 where @xmath151 , @xmath111 is the variance of @xmath80 , and @xmath152 is the real function corresponding to @xmath40 , and @xmath153 is the variance of @xmath154 , and finally , @xmath31 .
[ th : sec3 ] please refer to the appendix .
@xmath155 is decreasing with @xmath156 .
so , in order to increase @xmath157 , most of the variance @xmath111 should be distributed on @xmath158 which have lower @xmath156 ( i.e. operate on smaller blocks ) .
particularly , the lower bound is minimized by setting @xmath159 this recovers the result in @xcite .
we derived a relation between the dependency spectrum of a boolean function and its correlation preserving properties .
this can be used in a variety of disciplines .
for example , in communication problems , cooperation among different nodes in a network requires correlated outputs which can be linked to the dependency spectrum through the results derived here . on the other hand ,
there are restrictions on the dependency spectrum based on the rate - distortion requirements ( better performance requires larger effective lengths ) .
we investigate this in @xcite , and show that the large blocklength single - letter coding strategies used in networks are sub - optimal in various problems .
we derived a new bound on the maximum correlation between boolean functions operating on pairs of sequences of random variable .
the bound was presented as a function of the dependency spectrum of the functions .
we developed a new mathematical apparatus for analyzing boolean functions , provided formulas for decomposing the boolean function into additive components , and for calculating the dependency spectrum of these functions .
the new bound has wide ranging applications in security , control and information theory .
the uniqueness of such a decomposition follows from the isomorphy relation stated in equation .
we prove that the @xmath80 given in the lemma are indeed the decomposition into the components of the direct sum .
equivalently , we show that @xmath160 @xmath78 , and @xmath161 @xmath162 .
first we check the equality @xmath78 .
let @xmath163 denote the n - length vector whose elements are all ones .
we have : @xmath164 where in ( a ) we have used 1 ) @xmath165 and 2 ) for any function @xmath166 of @xmath167 , @xmath168 , and ( b ) holds since @xmath169 . .
it remains to show that @xmath162 .
the next proposition provides a means to verify this property .
fix @xmath170 , define @xmath171 , and @xmath172 .
@xmath166 is an element of @xmath173 if and only if @xmath174 it is constant in all @xmath175 , @xmath176 , and @xmath177 it has @xmath125 mean on all @xmath175 , when @xmath178 . [
prop : belong1 ] by definition , any element of @xmath173 satisfies the conditions in the proposition .
conversely , we show that any function satisfying the conditions @xmath174 and @xmath177 is in the tensor product .
let @xmath179 .
assume @xmath180 for some @xmath181 $ ] .
then : @xmath182 where we have used linearity of expectation in ( a ) , and the last two equalities use the fact that @xmath183 which means it satisfies properties @xmath174 and @xmath177 .
so far we have shown that @xmath184 .
now assume @xmath185 .
then : @xmath186 so , @xmath187 .
by assumption we have @xmath188 .
\1 ) for two n - length binary vectors @xmath19 , and @xmath190 , we write @xmath191 if @xmath192 $ ] .
the set @xmath193 equipped with @xmath194 is a well - founded set ( i.e. any subset of @xmath193 has at least one minimal element ) .
the following presents the principle of noetherian induction on well - founded sets : let @xmath195 be a well - founded set . to prove the property @xmath196 is true for all elements @xmath197 in @xmath198 , it is sufficient to prove the following + 1 ) * induction basis : * @xmath196 is true for all minimal elements in @xmath198 .
+ 2 ) * induction step : * for any non - minimal element @xmath197 in @xmath198 , if @xmath199 is true for all minimal @xmath200 such that @xmath201 , then it is true for @xmath197
. we will use noetherian induction to prove the result .
let @xmath202 $ ] be the @xmath203th element of the standard basis
. then @xmath204 . by the smoothing property of expectation , @xmath205 .
assume that @xmath206 , @xmath207 .
then , @xmath208 + 2 ) this statement is also proved by induction .
@xmath209 is a function of @xmath21 , so by induction @xmath210 is also a function of @xmath21 .
+ 3 ) let @xmath211 $ ] be defined as the @xmath83th element of the standard basis , and take @xmath212 , j\neq j'$ ] .
we have : @xmath213 where we have used the memoryless property of the source in ( a ) and ( b ) results from the smoothing property of expectation .
we extend the argument by noetherian induction .
fix @xmath214 .
assume that @xmath215 , and @xmath216 .
@xmath217 the second and third terms in the above expression can be simplified as follows .
first , note that : @xmath218 our goal is to simplify @xmath219 .
we proceed by considering two different cases : + * case 1 : * @xmath220 and @xmath221 : replacing the terms in the original equality we get : @xmath226 where in ( b ) we have used that @xmath80 s are uncorrelated , and ( a ) is proved below : @xmath227 + * case 2 : * assume @xmath228 : @xmath229 + * case 3 : * when @xmath230 the proof is similar to case 2 .
+ 4 ) clearly when @xmath231 , the claim holds .
assume it is true for all @xmath190 such that @xmath232 .
take @xmath170 and @xmath233 , i_t=1 $ ] arbitrarily .
we first prove the claim for @xmath234 : @xmath235 where in ( a ) we have used @xmath236 , also ( b ) follows from @xmath237 , ( c ) uses @xmath238 , and finally , ( d ) uses the induction assumption .
now we extend the result to general @xmath239 .
fix @xmath240 .
assume the claim is true for all @xmath190 such that @xmath241 ( i.e @xmath242 ) .
we have : @xmath243 the proof involves three main steps .
first , we bound the pearson correlation between the real - valued functions @xmath45 , and @xmath166 . in the second step ,
we relate the correlation to the probability that the two functions are equal and derive the lower bound .
finally , in the third step we use the lower bound proved in the first two steps to derive the upper bound . + * step 1 : * from remark [ rem : exp_0 ] , the expectation of both functions is 0 .
so , the pearson correlation is given by @xmath244 .
our goal is to bound this value .
we have : @xmath245 in ( a ) we have used remark [ rem : dec ] , and in ( b ) we use linearity of expectation .
using the fact that @xmath79 and lemma [ lem : tensor_dec ] , we have : @xmath246 we replace @xmath80 and @xmath247 in : @xmath248 @xmath249 in ( a ) we have used the fact that in a pair of dms s , @xmath114 and @xmath250 are independent for @xmath251 .
( b ) holds since from proposition [ prop : belong2 ] , @xmath252 $ ] .
we prove ( c ) in lemma [ lem : init ] below . in ( d ) we have used proposition [ pr : partfun ] .
using equations and we get : @xmath256 + * step 2 : * we use the results from step one to derive a bound on @xmath257 .
define @xmath258 , @xmath259 , @xmath260 , and @xmath261 , then @xmath262 we write this equation in terms of @xmath263 , q , and r using the following relations : @xmath264 solving the above we get : @xmath265 we replace @xmath266 , and @xmath267 in by their values in @xmath268 : @xmath269 on the other hand @xmath270 , where the last equality follows from the fact that @xmath80 s are uncorrelated .
this proves the lower bound .
next we use the lower bound to derive the upper bound . +
* step 3 : * the upper - bound can be derived by considering the function @xmath271 to be the complement of @xmath148 ( i.e. @xmath272 . ) in this case @xmath273 .
the corresponding real function for @xmath271 is : @xmath274 so , @xmath275 .
using the same method as in the previous step , we have : @xmath276 on the other hand @xmath277 .
so , @xmath278 this completes the proof .
h. s. witsenhausen , `` on sequences of pair of dependent random variables , '' _ siam journal of applied mathematics _ ,
28 , no . 1 ,
pp . 100113 , 1975 . f. s. chaharsooghi , a. g. sahebi , and s. s. pradhan , `` distributed source coding in absence of common components , '' in _ information theory proceedings ( isit ) , 2013 ieee international symposium on _ , july 2013 , pp .
13621366 .
a. mahajan , a. nayyar , and d. teneketzis , `` identifying tractable decentralized control problems on the basis of information structure , '' in _ 2008 46th annual allerton conference on communication , control , and computing _ , sept 2008 , pp .
14401449 .
m. reed and b. simon , _ methods of modern mathematical physics , i : functional analysis_.1em plus 0.5em minus 0.4emnew york : academic press inc . ltd . ,
1972 . f. shirani , m. heidari , s. s. pradhan , _ on the sub - optimality of single - letter coding in multi - letter communications _ , arxiv.org , jan 2017 .
f. shirani , s. s. pradhan , _ on the correlation between boolean functions of sequences of random variables _ , arxiv.org , jan 2017 . | in this paper , we establish a new inequality tying together the effective length and the maximum correlation between the outputs of an arbitrary pair of boolean functions which operate on two sequences of correlated random variables .
we derive a new upper - bound on the correlation between the outputs of these functions .
the upper - bound is useful in various disciplines which deal with common - information .
we build upon witsenhausen s @xcite bound on maximum - correlation .
the previous upper - bound did not take the effective length of the boolean functions into account .
one possible application of the new bound is to characterize the communication - cooperation tradeoff in multi - terminal communications . in this problem , there are lower - bounds on the effective length of the boolean functions due to the rate - distortion constraints in the problem , as well as lower bounds on the output correlation at different nodes due to the multi - terminal nature of the problem . |
the start of the lhc has greatly motivated detailed phenomenological studies of scenarios which involve physics beyond the standard model ( sm ) . among them
d - brane constructions in string theory are one of the most promising framework in which the sm can be embedded and extended . such brane constructions naturally lead to extra anomalous @xmath0 s in the four dimensional low energy theory and , in turn , to the presence of possible heavy @xmath1 particles in the spectrum
. these particles should be among the early findings of lhc and besides for the above cited models they are also a prediction of many other theoretical models of the unification of forces ( see @xcite for a recent review ) . in
@xcite we have considered a minimal extension of the minimal supersymmetric standard model ( mssm ) with a single extra @xmath2 gauge symmetry in a string - inspired setup .
we believe that our model encodes the key features of the low - energy sector of some of those brane construction . in this framework we studied in @xcite the radiative decay of the next to lightest supersymmetric particle ( nlsp ) into the lightest supersymmetric particle ( lsp ) .
this kind of process is very interesting since it might be the first one where the lsp could be observed at lhc @xcite and at the upcoming ilc @xcite .
under suitable assumptions the lsp in our model turns out to be an axino @xcite , the fermion component of the stckelberg supermultiplet related to the anomaly cancellation mechanism ( see for details @xcite ) . without loss of generality
we assume a wino - like nlsp . in the following we
just give the interaction term which involve the axino and the wino relevant for our analysis .
the interaction term , written in terms of four components majorana spinors are in the weyl representation .
] , is given by & & i = _ w |_2 _ 5 [ ^,^](_a _ ) _ s where @xmath3 is the neutral wino , @xmath4 is the axino , @xmath5 is the photon , @xmath6 the weinberg angle , @xmath7 and @xmath8 respectively the @xmath2 and @xmath9 coupling constants , @xmath10 the @xmath11 anomaly factor and @xmath12 the @xmath13 mass .
the rate of the radiative decay ( @xmath14 ) is _
^(2 ) = g_2 ^ 4 ^2_w @xmath15 ^ 2 [ gamma ] where @xmath16 , while @xmath17 and @xmath18 are respectively the wino and axino masses . as we showed in @xcite ,
the radiative decay is the most dominant wino decay mode with a br close to 1 ( @xmath19 ) , so we can use ( [ gamma ] ) to give an estimation of the wino mean life time _ _ 2 [ meanlife ]
in order to fall into the wmap range in the most experimentally attractive situation , we considered a light lsp ( @xmath20 ) and a mass gap of order @xmath21 , which imply more energetic and therefore easier to detect photons .
this requirement is necessary because the detector resolution increases with energy , while at low energy there is an obstruction for the detection of photons due to bremsstrahlung , qcd background and absorption before the detection from the calorimeter @xcite .
+ moreover we considered a universal squark mass @xmath22 for the first two squark generations ( since under this assumption they are nearly degenerate ) and we assumed flavor blindness @xcite .
the contribution from the third generation squarks is always negligible . and the universal squark mass @xmath22 . ] in fig .
[ fig : winoprod ] we summarize the results obtained in @xcite by plotting the number of directly produced winos as a function of @xmath18 and @xmath23 having assumed 14 tev of center of mass energy and 100 fb@xmath24 of integrated luminosity .
since the br is almost close to one this is also the number of photons in the final state .
the number of photons produced is of the order of @xmath25 . in our analysis
we follow @xcite,@xcite-@xcite , where the nlsp decay in the gmsb framework is controlled by the parameter @xmath26 .
if the nlsp lifetime is not too long ( @xmath27 ) photons originate close to the primary interaction vertex ( `` prompt photons '' ) . in the case of large @xmath26 and
therefore long lived neutralinos the resulting photons are non - pointing . from now on we fix the axino mass @xmath28 gev and the universal squark mass @xmath29 tev . in our framework the role of @xmath26 is played by the ratio @xmath30 . in the following we discuss two different cases : short lived nlsp and long lived one .
we compare the number of photons produced by radiative decay with the ones produced by the cascade decays of all the other supersymmetric processes .
we slightly modified the herwig code 6.5 @xcite in order to take into account the new axino state in the neutral sector .
it should be stressed that herwig does not implement extra @xmath13 in a supersymmetric framework .
this in turn implies that the total number of photons can be underestimated due to the lack of sparticles interactions with the @xmath13 .
however this problem can be overcome by assuming a decoupled @xmath13 either because it is very heavy or because it is extra - weak coupled .
we generated by herwig 2-partons@xmath31 2-sparticles events , using about 1 @xmath32 of integrated luminosity but we have not considered the case of sm particles produced directly in the parton - parton interaction .
a good discriminant variable of the process is the @xmath33 of the photons produced by radiative decay , in particular in the region of @xmath33 between 30 - 80 gev / c . the corresponding distribution is shown in fig .
[ fig : pta3 ] .
we denote in red the number of @xmath34 s radiatively produced from the decay of the wino , in blue the number of @xmath34 s from all the other processes while in black the sum of the two .
we assumed @xmath35 , which is obtainable with @xmath36 tev and @xmath37 .
distribution of photons ( in log scale ) for @xmath38 susy events .
] we performed the same cut on the number of generated photons as in @xcite with @xmath39 gev and with pseudorapidity @xmath40 , @xmath41 , which provides a good way to further suppress the susy background . the result obtained by using herwig in generating @xmath42 net events
is given in fig .
[ fig : num_phot_comp_before ] . the most important difference between our case and the gmsb1 sample
@xcite is in the number of events with zero or two photons in the final state .
the latter in particular is only 30 in our case .
this behavior can be related to the squark masses we have considered . in our case
they are about 3.5 tev , while in the gmsb1 they are lower than 1 tev ( @xmath43 900 gev ) .
we choose the value of 3.5 tev for the squark masses since in this case the number of directly produced winos essentially depends only on @xmath18 ( see fig .
[ fig : winoprod ] ) .
the number of produced squarks is low since they have a high mass .
hence they give a lower contribution to the nlsp production .
if we consider lighter squarks , with masses less than 1 tev , there is an increasing in the number of events with 2 photons in the final state . in any case
the channel with one photon in the final state is always the dominant one .
+ the key point in the analysis is the sm background discrimination .
we considered the same sm background as in @xcite : events with qcd jets , single gauge bosons ( w and z ) production , di - boson , @xmath44 and @xmath45 production . in order to disentangle the sm background from the signal we require a standard preselection cut for susy - like signatures : *
at least four jets must be present with @xmath46 gev ( @xmath47 gev for the leading jet ) ; * missing transverse energy @xmath48 gev .
after having applied these preselection cut we are able to reconstruct photons with @xmath49 gev and @xmath40 , @xmath41 .
the jet - finder algorithm used in our analysis is the durham - type clustering ktclus @xcite .
the cut on the @xmath50 requires some care . in hadron colliders , the initial momentum of the colliding partons along the beam axis
is not known since the energy of each hadron is distributed and constantly exchanged between the partons .
hence the total amount of missing energy can not be determined in a straightforward way .
however the initial energy of particles travelling transverse to the beam axis is zero and thus any net momentum in the transverse direction denotes missing transverse energy . to determine the latter for the i - th not - detected particle in each generated event
, we considered the following procedure : take a transverse direction ( perpendicular to the z axis ) and define the vector @xmath51 whose direction is given by its momentum . then @xmath52 ^ 2+[(e^{miss}_{t})^{tot}_y]^2}$ ] where @xmath53 and analogously for the @xmath54 component . the result is shown in fig .
[ fig : smbackground ] . in this plot
the number of photons from the nlsp decay is that generated by herwig while the number of background photons is the reconstructed one @xcite .
we note that in the channels with @xmath55 the signal can be easily disentangled by the sm background .
is the distance traveled by @xmath56 before its radiative decay .
@xmath57 is the value of the displaced vertex . ] in the case in which the wino is a long lived particle the reconstruction of the emitted photon after its decaying plays an important role .
if the nlsp has a significant decay length , a photon will not point back " to the primary interaction point ( @xmath58 in fig .
[ fig : nopoint ] ) but towards a kind of `` virtual '' point @xmath59 .
the relevant discriminant variable is @xmath57 , the distance between @xmath58 and @xmath60 . using simple trigonometric relations we have : z _ = d_z - d_xywith @xmath61 , the ratio between the photon momentum along the beam direction ( z - axis ) and the transverse momentum . by fitting the vertex resolution along z as a function of the photon energy for @xmath62 ( see plot in section electrons and photons " in @xcite ) we derive the functional dependence of the standard deviation @xmath63 for the gaussian distribution associated to @xmath57 on the energy in gev of all the photons as 0.7 + [ eq : sigma ] to each @xmath57 we can associate a random number sampled from a gaussian distribution centered on that specific value of @xmath57 and with @xmath63 given by the relation in ( [ eq : sigma ] ) .
we considered a wino mean life time @xmath64 s , so that the distance travelled by this particle before its radiative decay is always within the atlas tracker radius ( @xmath43 115 cm ) . in fig .
[ fig : displaced ] we show the results of the analysis for photons of @xmath65 gev and with @xmath66 ( photons with great angle to the beam direction , because these are more easily detected than the ones along the z direction ) , for an integrated luminosity of 10 @xmath32 . by considering the value of their energy and the contribution of the tail of the associated gaussian distribution
, we observe a consequent enlargement of the distribution of @xmath34 s not coming from @xmath3 , which otherwise would be a spike centered on the origin of the x - axis in fig .
[ fig : displaced ] .
the red line stands for photons produced by radiative wino decay , while the black one from the others susy processes .
we can see that in the region @xmath67 cm the photons from the radiative decay are well distinguished from the background ones .
since we have assumed @xmath68 , @xmath69 gev and @xmath64 s , we get from ( [ meanlife ] ) [ ultima ] to satisfy ( [ ultima ] ) we need either a super - heavy @xmath13 ( @xmath70 tev if @xmath71 ) or an extra - weakly coupled ( @xmath72 if @xmath36 tev ) one .
this decoupling makes our simulation more consistent since in it we have neglected the @xmath13 contribution .
cases of extra - weak @xmath13 were already studied in the past @xcite . for photons of @xmath65 gev and @xmath73
: the red line stands for photons produced by wino decay , the black one for photons from the other susy processes . ]
the most important result we get is shown in fig . [
fig : displaced ] which gives a distinctive behavior of the photons coming from our model with respect to those coming from other susy processes .
the distribution of @xmath57 is centered in zero while the half - width is a function of the nlsp decay length @xmath74 .
this means that the radiative produced photons are preferably emitted along the nlsp direction or more generally at very small angle .
thus the signal can be disentangled from the background by applying a suitable cut in a region @xmath75 cm .
our result can be compared with @xcite in which the nlsp is a long lived wino - like particle in the particular gauge mediated model dubbed gmsb3 .
the main difference is in the mass gap . in our model
we considered a mass gap of @xmath76 and an lsp of 124 gev , while in the gmsb3 case the mass gap is much higher : the gravitino ( the lsp ) is almost massless ( @xmath77 gev ) and the nlsp has a mass @xmath78 gev . in this case
the energy distribution for the photons produced by the nlsp decay is peaked on the value of the nlsp mass @xmath79 gev while in our case the peak is around @xmath80 gev .
moreover in the gmsb3 @xmath81 m and so the nlsp long lived particles tend to escape the detector before their decay .
this implies a sizable reduction in the number of photons detected .
the authors gratefully acknowledge the atlas group of tor vergata , in particular prof .
a. di ciaccio , g. cattani and r. di nardo for many stimulating discussions and help .
a. r. would like to thank prof .
m. raidal and dr .
k. kannike for discussions and the esf jd164 contract for financial support .
a. lionetto and a. racioppi , nucl .
b * 831 * ( 2010 ) 329 [ arxiv:0905.4607 [ hep - ph ] ] .
h. baer and t. krupovnickas , jhep * 0209 * ( 2002 ) 038 [ arxiv : hep - ph/0208277 ] .
h. baer , a. mustafayev , e. k. park and x. tata , jhep * 0805 * ( 2008 ) 058 [ arxiv:0802.3384 [ hep - ph ] ] .
h. k. dreiner , o. kittel and u. langenfeld , phys .
d * 74 * ( 2006 ) 115010 [ arxiv : hep - ph/0610020 ] .
m. terwort [ atlas collaboration and cms collaboration ] , arxiv:0805.2524 [ hep - ex ]
. m. terwort , `` study of gmsb models with photon final states using the atlas detector '' , phd desy - thesis , 2009 .
g. corcella _ et al .
_ , jhep * 0101 * ( 2001 ) 010 [ arxiv : hep - ph/0011363 ] . | an extension of the standard model by at least one extra u(1 ) gauge symmetry has been investigated by many authors . in this paper
we explore the possibility that this extra u(1 ) is anomalous .
one of the possible signatures of this model could be given by the photons produced in the decays of the nlsp into the lsp . beyond the sm
; lhc ; anomalous u(1 ) models ; nlsp radiative decay |
to continuous introduction of mobile devices and services , future cellular systems are facing a significantly increased number of mobile devices requesting large data volume . to accommodate such a large growth of mobile devices , there are active researches on the 5th generation ( 5 g ) cellular system .
new targets for the 5 g cellular system are to support latency - sensitive applications such as tactile internet @xcite and low energy consumption for machine - type communication ( mtc ) @xcite or the internet of things ( iot ) @xcite .
unfortunately , a cellular system can not achieve the two targets simultaneously , but a non - trivial tradeoff can exist .
although this tradeoff is very important to 5 g cellular system designers , related researches are rare .
this is because it is often hard to deal with the latency and the energy consumption analytically so that intensive simulation - based network plannings are widely spread @xcite , @xcite .
however , this approach becomes impractical when the network components , such as the number of users and bs antennas are scaled up .
more viable approach is to analyze the network .
this paper mainly concentrates on the analysis about the tradeoff between the latency and the energy consumption in a promising 5 g cellular system . in 5 g cellular systems ,
there has been great interest to a large - scale antenna system ( lsas ) , a.k.a .
massive multiple - input multiple - output ( mimo ) , in which very large number of antennas are equipped at a base station ( bs ) to serve many users simultaneously @xcite .
its inherent merits come from massive spatial dimensions , which include i ) generating sharp beams for intended users to improve spectral efficiency by suppressing unintended interference @xcite , @xcite , ii ) reducing transmit energy while guaranteeing quality of service ( qos ) @xcite , and iii ) allowing a complexity - efficient transceiver algorithm @xcite . in order to achieve such advantages ,
an appropriate channel state information ( csi ) acquisition process is essential . to acquire csi ,
a widely - accepted approach is the training - based transmission in which a frame is divided into two phases : one is the training phase , in which users transmit known training signals and the bs estimates the csi , and the other is the data transmission phase , in which the users transmit information - bearing signals and the bs extracts the information by utilizing the estimated csi .
even if the training - based transmission is not optimal in information - theoretic point of view , it gives an efficient way to acquire the csi as well as to provide the optimal degrees of freedom in the high signal - to - noise ratio ( snr ) regime @xcite . in order to analyze the latency in the training - based lsas ,
it is necessary to optimize the user scheduling policy as well as the resource allocation under reasonable and practical constraints .
if this optimization is not performed , it often gives an inappropriate cellular system design .
the optimization of the training - based transmission is firstly investigated by hassibi and hochwald @xcite .
they consider the mimo point - to - point channel with a capacity - approaching transmitter / receiver pair and successfully derive the optimal system parameters as a function of snr and other parameters .
later , this results are extended to the mimo broadcast channel @xcite , multiple access channel @xcite , relay channel @xcite , and interference channel @xcite .
however , these works optimize the energy and time dedicated to the training phase only under a given user set so that it can not be directly applied to the latency - energy tradeoff in the lsas . in order to evaluate the latency of the lsas ,
it is necessary to further optimize those variables under the optimal scheduling policy .
the scheduling policies to minimize the latency ( or delay ) under a minimum rate constraint or to maximize spectral efficiency under a maximum latency constraint have been widely investigated in literature under various system models . in @xcite
, the system average delay is optimized by using combined energy / rate control under average symbol - energy constraints . in @xcite ,
delay - optimal energy and subcarrier allocation is proposed for orthogonal frequency division multiple access ( ofdma ) . in @xcite ,
the energy minimizing scheduler , by adapting energy and rate based on the queue and channel state is proposed .
however , most of them assume perfect csi at transmitter and receiver so that it often overestimates the network - wise performance .
also , their scheduling policies are too complicated to be analyzed for providing an intuitive insight on the network - wise performance .
thus , a practically optimal scheduling policy for the training - based lsas is needed and an intuitive analysis is desired to provide an insight on the latency - energy tradeoff in the lsas .
decreasing the latency in the lsas is closely related to increasing the spectral efficiency , because higher spectral efficiency results in a smaller transmission completion time if the number of users and their rate constraints are given .
in addition , the spectral efficiency of a multiple - access channel with @xmath1 bs antennas and @xmath10 scheduled users is asymptotically expressed as @xmath11 as @xmath12 , which implies that the spectral efficiency can be enhanced by scheduling users as many as possible in the lsas .
however , most literature assumes that _ orthogonal _ pilots are allocated to users so that the maximum number of scheduled users is limited by the number of available pilots in practice .
actually , there is no reason that orthogonal pilots are optimal for the latency - energy tradeoff so that it is natural to consider _ non - orthogonal _ pilots in general .
there are a few results related to the case using non - orthogonal pilots . in @xcite ,
optimal non - orthogonal pilots for minimizing channel estimation error are derived and it turns out that finding the optimal non - orthogonal pilots is equivalent to solving the grassmannian subspace packing problem . in @xcite ,
an iterative algorithm is proposed to find optimal non - orthogonal pilots for maximizing the number of users with a minimum rate constraint in a downlink lsas .
however , they still do not address the effect of the non - orthogonal pilots on the latency and it would be very interesting to find whether the use of non - orthogonal pilots can reduce the latency and when and how much reduction can be obtained over the case of using orthogonal pilots . in this paper , we are interested in an uplink training - based lsas serving many users with an average energy constraint , in which each user has a limited average energy for transmitting a frame . in addition , we assume a block rayleigh fading model and a practical receiver such as the maximum ratio combining ( mrc ) or the zero - forcing ( zf ) receiver and focus on the resource allocation and multiple access strategy ( to be specific , pilot allocation , user grouping and scheduling , and energy allocation ) . the main target of this paper is to address the following question : * _ how much time is needed for guaranteeing the minimum throughput to all users in the uplink training - based lsas ? _ * it is , in general , hard to address this question analytically so that we look into the two asymptotic regimes , high and low energy regimes and successfully derive the effect on the network latency according to the energy consumption . the main contributions of this paper are summarized as follows : * we optimize the uplink scheduling policy for minimizing the latency with guaranteeing the minimum rate constraint .
the optimizing variables are the scheduling groups in which users are simultaneously scheduled , the scheduling portion indicating how often each scheduling group actually transmits , and the energy allocation indicating how much portion of energy is dedicated to the training phase .
this problem is transformed into an equivalent problem of maximizing the spectral efficiency with the rate constraint .
the optimal scheduling policy is obtained by solving the binary integer programming ( bip ) and it is proved that the optimal solution of the original bip can be obtained by a linear programming relaxation with a polynomial - time complexity . * we investigate the asymptotic performance of the proposed optimal uplink scheduling policy for a large number of users .
we derive a simple close - form expression for the asymptotic network latency and find the optimal parameters for the proposed optimal uplink scheduling policy . then , we identify four operating regimes of the training - based lsas according to the growth or decay rate of the average received signal quality , @xmath0 , and the number of bs antennas , @xmath1 .
it turns out that orthogonal pilot sequences are optimal only when the average received signal quality is sufficiently good and the number of bs antennas is not - so - large . in other regimes , it turns out that non - orthogonal pilot sequences become optimal .
in fact , the use of non - orthogonal pilots can reduce the network latency by a factor of @xmath6 when the received signal quality is quite poor ( @xmath4 ) or by a factor of @xmath7 when the received signal quality is sufficiently good ( @xmath8 ) and the number of bs antennas is sufficiently large ( @xmath13 ) .
the remainder of this paper is organized as follows . in section
ii , a detailed model description is presented including channel , energy , and signal models for a training - based lsas . in section iii ,
the uplink scheduling policy problem is formulated and its optimal solution is provided .
numerical experiments to verify the superiority of the proposed uplink scheduling policy are shown in section vi . in section
v , an asymptotic analysis provides the closed - form network latency and optimal parameters for the proposed uplink scheduling policy .
finally , conclusion is drawn in section vi .
matrices and vectors are respectively denoted by boldface uppercase and lowercase characters . also , @xmath14 , @xmath15 , and @xmath16 stand for the transpose , conjugate transpose , and cardinality of a set , respectively , and @xmath17 and @xmath18 are the natural logarithm and the logarithm with base 2 , respectively .
also , @xmath19 denotes the function rounding @xmath20 towards the nearest integer , @xmath21 , and @xmath22 denotes the indicator function .
@xmath23 denotes the distribution of a circularly symmetric complex gaussian random vector with mean vector @xmath24 and covariance matrix @xmath25 and @xmath26 $ ] and @xmath27 denotes the statistical expectation and the statistical variance , respectively .
finally , standard order notations in @xcite are used . for better readability ,
frequently used symbols are summarized in table i. ?
symbols & descriptions + @xmath1 & # of bs antennas + @xmath28 & # of total users and their index set , @xmath29 + @xmath30 & # of frames in a uplink scheduling period and its minimum with guaranteeing @xmath31 bits to all of users .
+ @xmath32 & network latency , optimized network latency and optimized asymptotic network latency normalized by @xmath33 with guaranteeing @xmath31 bits to all of users .
+ @xmath34 & # of sub - frames in a frame + @xmath35 & # of symbols in a sub - frame + @xmath36 & # of symbols in the training phase or the data transmission phase + @xmath37 & bandwidth and the symbol duration + @xmath38 & bandwidth inefficiency with @xmath39 + @xmath40 & the average energy constraint of user @xmath41 .
+ @xmath42 , { r}_k[f;t ] $ ] & the achievable rate of users @xmath41 in sub - frame @xmath43 of frame @xmath44 and its approximation ( bits / hz ) + @xmath31 & throughput threshold ( bits ) + @xmath45 & the scheduling group and its scheduling portion + @xmath46 & the energy dedicated to the training symbols or data symbols ( joule / symbol ) +
we consider an uplink lsas consisting of a bs with @xmath1 antennas , and @xmath33 single - antenna users as illustrated in fig .
it is assumed that the users are randomly distributed on the cell coverage region and they want the quality of service ( qos ) on their own data traffic ( rate , latency , and reliability ) so that the bs serves these users persistently and try to guarantee their qos . a two - phase frame structure with training and data transmission phases , illustrated as in fig .
2 , is used . for every uplink scheduling period ,
the bs broadcasts the scheduling information and then the users transmit @xmath47 frames in uplink direction step by step .
the frame of time length @xmath48 seconds and bandwidth @xmath49 hz is divided into @xmath34 equal - bandwidth sub - frames by partitioning frequency domain by using the orthogonal division multiplex access ( ofdma ) , single - carrier frequency domain multiple access ( sc - fdma ) or any good one of the newly considered waveforms @xcite .
the sub - frame consists of @xmath35 symbols of time period @xmath50 seconds . in the training phase of time period
@xmath51 seconds , the scheduled users send @xmath52 training symbols , and the bs estimates the uplink channels .
then , in the data transmission phase of the remaining time period @xmath53 seconds , all of the scheduled users transmit @xmath54 data symbols to the bs simultaneously in a space division multiple access ( sdma ) manner .
sub - frames and @xmath35 symbols in the sub - frame can be arbitrarily configured both in time and frequency domains .
for example , in the long - term evolution ( lte ) , one sub - frame includes @xmath55 symbols ( 14 symbols in time domain and 12 sub - carriers in frequency domain ) and one frame includes @xmath56 sub - frames in time domain .
] so , the transmit signal vector of user @xmath41 , who is allocated to the @xmath43th sub - frame in frame @xmath44 , is written as @xmath57 = { \left [ { { { \left ( { { \bf{x}}_j^{{\rm{tr}}}[f;t ] } \right)}^h},{{\left ( { { \bf{x}}_j^{{\rm{dt}}}[f;t ] } \right)}^h } } \right]^h},\ ] ] where @xmath58 $ ] is the @xmath59 training symbol vector for the training phase and @xmath60 $ ] is the @xmath61 data symbol vector for the data transmission phase . in every sub - frame , at most @xmath1 users are scheduled and the set of users scheduled in sub - frame @xmath43 of frame @xmath44 is denoted as @xmath62 $ ] .
assume that @xmath63|=k$ ] .
within one block of @xmath35 symbols , the @xmath64 received signal matrix , denoted as @xmath65=\left[\mathbf{y}_1[f;t],\mathbf{y}_2[f;t],\dots,\mathbf{y}_n[f;t]\right]$ ] , is given as @xmath66 & = \mathbf{g}_{\mathcal{s}[f;t]}[f;t]\mathbf{x}^h[f;t ] + \mathbf{v}[f;t ] , \end{split}\ ] ] where @xmath67}[f;t ] = \left[\mathbf{g}_{j}[f;t]\right]_{j\in\mathcal{s}[f;t]}$ ] is the @xmath68 channel matrix , @xmath69=[\mathbf{x}_1[f;t],\dots,\mathbf{x}_k[f;t ] ] $ ] is the @xmath70 transmitted signal matrix and @xmath71=\left[\mathbf{v}_{1}[f;t],\mathbf{v}_{2}[f;t],\dots,\mathbf{v}_{n}[f;t]\right]$ ] is the @xmath64 noise matrix , whose elements are independent and identically distributed ( i.i.d . ) random variables with @xmath72 .
the @xmath73 flat - fading channel vector between the bs and user @xmath41 at the sub - frame @xmath43 of frame @xmath44 , @xmath74 $ ] , can be written as @xmath75 = \sqrt { { \beta _ { j } } } { { \mathbf{h}}_{j}}[f;t],\ ] ] where @xmath76 \in { \mathbb{c}^{m \times 1}}$ ] is the short - term csi whose elements are i.i.d .
random variables with @xmath72 and @xmath77 is the long - term csi depending on the path - loss and shadowing .
the long - term csi between the bs and user @xmath41 is modeled as @xmath78 , where @xmath79 is the wireless channel path - loss exponent and @xmath80 is the distance between the bs and user @xmath41 .
we assume that rayleigh block fading model , where the short - term csi of each user remains constant within a given frame but is independent across different frames , while the long - term csi does not vary during a much longer interval .
further , it is assumed that the long - term csi of all users is perfectly known at the bs . since each of users has a different limited battery capacity , recharge process , or radio frequency ( rf ) transmitter , they are assumed to be limited to spend energy for transmitting each sub - frame .
let @xmath81 be the average allowed energy level ( in joule ) of user @xmath41 per sub - frame of length @xmath48 .
the energy is consumed during both the training phase of length @xmath82 and the data transmission phase of length @xmath83 .
so , the consumed energy transmitting each sub - frame needs to meet @xmath84\|^2 = \mathbb{e}\|\mathbf{x}_j^{\rm{tr}}[f;t]\|^2+\mathbb{e}\|\mathbf{x}_j^{\rm{dt}}[f;t]\|^2\le e_j,~\forall j.\ ] ] letting @xmath85\|^2/l$ ] and @xmath86\|^2/(n - l)$ ] be the transmit energy of each training symbol and data symbol , respectively , the constraint is represented as @xmath87 in the sequel , we drop the sub - frame index @xmath43 and the frame index @xmath44 if there is no ambiguity . to estimate the csi of @xmath10 scheduled users , the bs allocates @xmath10 pilot sequences with length of @xmath52 .
let @xmath88 $ ] be the @xmath89 pilot matrix with normalized columns , i.e. , @xmath90 for all @xmath41 .
the training symbol vector of scheduled user @xmath41 during the training phase is @xmath91 . for equalizing the difference of all users channel estimation quality ( maximizing the worst ) ,
the _ channel - inversely power - controlled pilots _ are assumed similarly as in @xcite , in which the average received signal energy of the users in @xmath62 $ ] is set to the common target received energy , @xmath92 .
so , the transmit energy at the training phase is set by @xmath93 then , the @xmath94 received signal matrix in frame @xmath44 at the bs during the training phase , denoted as @xmath95 $ ] , can be written as @xmath96}\boldsymbol\psi _
{ l , k}^h + { { \mathbf{v}}}^{\rm{tr}},\end{aligned}\ ] ] where @xmath97 } = \left[\mathbf{h}_{j}\right]_{j\in\mathcal{s}[f;t]}$ ] , and @xmath98 $ ] is the noise vector during the training phase . using the minimum mean - square error ( mmse ) channel estimator @xcite , the estimated short - term csi of user @xmath41 , denoted as @xmath99 , can be written as @xmath100 denote the channel estimation error by @xmath101 .
the following lemma informs the property of the mmse channel estimation . with the channel - inversely power - controlled pilots , @xmath102 and @xmath103
are mutually independent and the channel estimation error variance @xmath104 is given by @xmath105 where @xmath106 are the eigenvalues of @xmath107 .
see the proof of theorem 1 in @xcite . during the data transmission phase ,
the @xmath108th received signal vector at sub - frame @xmath43 of frame @xmath44 , @xmath109 , is given by @xmath110 = \sum\limits_{j\in\mathcal{s}[f;t ] } \sqrt{p_j^{\rm{dt}}\beta_j}\mathbf{h}_j s_{jl}[f;t ] + \mathbf{v}_l,\ ] ] where @xmath111\sim\mathcal{cn}(0,1)$ ] is the @xmath108th information - bearing data symbol of user @xmath41 at sub - frame @xmath43 of frame @xmath44 . with
the channel - inversely power - controlled non - orthogonal pilots , the achievable rate of user @xmath122 of the mrc or zf receiver is approximated by @xmath127 \approx \left\ { { \begin{array}{*{20}{l } } { \log_2\left ( { 1 + \gamma _ k[f;t ] } \right),}&{{\text{if } } k \in { \cal s}[f;t],}\\ { 0,}&{{\text{if } } k \notin { \cal s}[f;t ] , } \end{array } } \right.\ ] ] where @xmath128 $ ] is given as @xmath129= \left\ { { \begin{array}{*{20}{l } } { \frac{{\left ( { 1 - \sigma _ { { \rm{tr}}}^2 } \right)\left
( { m - k } \right)p_k^{{\rm{dt}}}{\beta _ k}}}{{1 + \sigma _ { { \rm{tr}}}^2\sum\limits_{j \in { \cal s}[f;t ] } { p_j^{{\rm{dt}}}{\beta _ j } } } } } , & { { \text{if zf}}},\\ { \frac{{\left ( { 1 - \sigma _ { { \rm{tr}}}^2 } \right)\left ( { m - 1 } \right)p_k^{{\rm{dt}}}{\beta
_ k}}}{{1 + { \sum\limits _ { j \in
\mathcal{s}[f;t]}}p_j^{{\rm{dt}}}{\beta _ j } - \left ( { 1 - \sigma _ { { \rm{tr}}}^2 } \right)p_k^{{\rm{dt}}}{\beta _ k}}}},&{{\text{if mrc}}}. \end{array } } \right . \end{split}\ ] ] the proof is similar as in @xcite , so omitted for brevity .
note that ( [ eq14 ] ) is valid only when @xmath130 for the zf receiver ( @xmath131 for the mrc receiver ) .
in fact , when such an approximation is used as the utility function of a scheduler , it can not select @xmath132 users , which under - utilizes the resource , especially when @xmath1 is small .
however , this paper deals with the lsas so that this problem becomes less significant .
additionally , to overcome this problem for a small @xmath1 , @xmath133 can be replaced by @xmath134 , which results in a small approximation error .
( similar approximation is shown in theorem 5 in @xcite . ) furthermore , this approximated version of the achievable rate becomes more accurate as the number of bs antennas increases .
[ rate achievability ] since the low latency communications are of interest , codewords can not be spread sufficiently in time domain .
also , the transmitter does not acquire its instantaneous csi , the transmission strategy can not be adapted according to the channel realization . in this paper
, we assume that user @xmath41 transmits information - bearing data symbols with a fixed rate of @xmath135 regardless of the channel realization . the bs can successfully receive this data symbol if and only if @xmath136 . from the strong law of large number
, we can show that @xmath136 almost surely as @xmath137 so that the rate @xmath135 is achievable in the lsas with high probability .
this is called _ channel hardening effect _ of the lsas , which implies that the instantaneous rate is approaching a deterministic quantity , i.e. , its mean , as @xmath138 .
see @xcite for more rigorous discussions . to maximize the average achievable rate ,
we need to design pilot sequences .
optimal pilot sequences @xmath139 for maximizing @xmath135 satisfies @xmath140 since the case @xmath141 is trivial so we omit .
let @xmath142 .
the channel estimation error variance ( [ eq7 ] ) is simplified into @xmath143 obviously , @xmath144 is a decreasing function with @xmath104 within @xmath145 .
so , the maximization of @xmath146 is equivalent to the maximization @xmath147 under @xmath148 , which is obtained when the eigenvalues are the same , i.e. , @xmath149 .
note that the above condition is identical to the condition minimizing the channel estimation error @xcite .
in fact , this condition is known as the welch bound equality ( wbe ) @xcite .
so , from lemma 3 , the channel estimation error variance is minimized at @xmath150 and the sinr is maximized at @xmath151 = \left\ { { \begin{array}{*{20}{l } } { \frac{{l\left ( { m - k } \right)p_k^{{\rm{dt}}}{\beta _ k}\overline { { p^{{\rm{tr } } } } } } } { { \left ( { 1 + { { \left ( { k - l } \right)}^ + } \overline { { p^{{\rm{tr } } } } } } \right)\left ( { 1 + \sum\limits_{j \in \mathcal{s}[f;t ] } { p_j^{{\rm{dt}}}{\beta _ j } } } \right ) + l\overline { { p^{{\rm{tr } } } } } } } , } & { { \text{if zf}},}\\ { \frac{{l\left ( { m - 1 } \right)p_k^{{\rm{dt}}}{\beta _ k}\overline { { p^{{\rm{tr } } } } } } } { { \left ( { 1 + \max
\left\ { { l , k } \right\}\overline { { p^{{\rm{tr } } } } } } \right)\left ( { 1 + \sum\limits_{j \in \mathcal{s}[f;t ] } { p_j^{{\rm{dt}}}{\beta _ j } } } \right)-l\overline{p^{\rm{tr}}}p_k^{\rm{dt}}\beta_k}},}&{{\text{if mrc}}. } \end{array } } \right . \end{split}\ ] ] if the pilot sequences are under - utilized ( orthogonal pilots are used ) , i.e. , @xmath152 , @xmath104 is simplified into @xmath153 , which means that the channel estimation error depends only on the energy dedicated to the training phase @xmath154 . however , if the pilot sequences are over - utilized ( non - orthogonal pilots are used ) , i.e. , @xmath155 , additional interference @xmath156 occurs .
[ optimal non - orthogonal pilots ] obviously , if @xmath152 , the optimal pilot sequences can be obtained from arbitrarily chosen @xmath10 columns of an @xmath157 unitary matrix . in the case of @xmath155 ,
one option is to over - sample an @xmath157 unitary matrix . although there are infinitely many sequences holding the wbe , one simple example is the discrete fourier transform ( dft ) based sequences , obtained as @xmath158,\ ] ] where @xmath159 and @xmath160 are arbitrarily chosen integers . note that the dft - based sequences are widely used in the design of the unitary - space time modulation @xcite or the feedback codebook @xcite .
the accuracy of the approximated rate in ( [ eq13 ] ) and ( [ eq17 ] ) is presented in fig .
3 , by setting @xmath161 , @xmath162 , @xmath163 , @xmath164db , @xmath165 , @xmath166 , and @xmath167 , @xmath168 . in fig .
3 , we plot the exact rate in ( [ eq12 ] ) and the approximated rate in ( [ eq13 ] ) according to the number of bs antennas , @xmath1 .
the dashed line represents the average of the exact rate and each error bar represents its standard deviation .
we can easily notice that the approximated rate is quite well fitted even at a small number of antennas in a wide range of average allowed energy level @xmath169 . ) and
( [ eq17 ] ) .
the line marked by a circle represents the results using the zf receiver and the line marked by a square represents the results using the mrc receiver . ]
according to the level of accessible information , uplink scheduling policies can be classified into two types .
a _ dynamic _ user scheduling , which is based on the _ instantaneous _ csi , can provide a substantial rate gain primarily because it allows the bs to select a subset of users whose channels are nearly orthogonal and the achievable uplink performance increases with the number of users .
however , obtaining the instantaneous csi at a bs will incur a large amount of uplink training resource cost .
in fact , due to the nature of the limited channel coherence time , it is hard for every user to participate in the scheduling pool without incurring non - negligible overhead . on the contrary ,
uplink scheduling exploits the long - term information only , such as the csi statistics and/or the average allowed energy levels , which can be easily acquired via infrequent feedback with negligible overhead . a static scheduling policy for @xmath47 frames
is defined as @xmath170 , where @xmath171 denotes the set of scheduling groups , @xmath172 denotes the set of scheduling portions , @xmath173 denotes the set of energy allocations for all users , and @xmath52 determines the sub - frame configuration . due to the nature of the static uplink scheduling , the output of the static scheduling policy should satisfy the following two constraints : 1 ) @xmath174 if @xmath175 , and 2 ) @xmath176 , where @xmath177 .
also , the scheduling portion is defined as @xmath178 = \mathcal{o}_q\}},\ ] ] so that it must satisfy @xmath179 . note that @xmath180 is the portion of the sub - frames allocated to scheduling group @xmath181 , @xmath182 , during one scheduling period consisting of @xmath183 sub - frames . in fact , since @xmath184 is not an integer , @xmath185 sub - frames may be allocated .
we assume that @xmath186 is sufficiently large so that the error @xmath187 is negligibly small . in general
, a user tries to send some of information to the bs via a wireless medium within various constraints .
however in cellular systems , the limited wireless medium is shared so that severe delay often occurs for waiting transmission turns and also for transmitting the target data volumes .
thus , to guarantee a low latency is one of the main hurdles to be addressed in future cellular systems .
the network latency ( delay ) is defined as @xmath188 ( sec ) , where @xmath189 is the minimum number of frames required to deliver the target throughput of @xmath31 bits for each user .
our definition is different to @xcite , @xcite , in which the delay is defined as the scheduling delay ( the waiting time for transmission turns ) only and the delay for transmitting the target throughput volume is ignored , and also to @xcite , @xcite , in which the delay is defined as the transmission completion time for the target throughput only and ignores the effect of the scheduling delay .
our definition includes both scheduling delay and transmission completion time . to minimize the latency under a given throughput constraint @xmath31 ,
it is sufficient to minimize the number of frames , @xmath47 .
thus , the optimization problem can be constructed as follows : [ eq19 ] align ( ) & s , & + & ( n - l)_t=1^s _
f=1^f r_j[f;t]_th , j , & + & p_j^tr + ( 1-)p_j^dt , p_j^tr , p_j^dt0 , j , & + & _ p _ q = , _
p=1^q_p= , & + & _ q=1^q d_q = 1 , d_q 0 , q , & + & l\{1,2, ,n-1}. & the constraint ( [ eq19]b ) is to guarantee the required throughput @xmath31 for each of users , ( [ eq19]c ) is to meet the average energy constraint , and ( [ eq19]d)-([eq19]e ) are to meet the static scheduler condition .
the variables in the optimization problem ( p ) are as follows : * which of users are simultaneously scheduled ? ( @xmath190 ) * how many sub - frames are allocated to each group ? ( @xmath172 ) * how the energy is allocated to the training and data transmission phases ?
( @xmath191 for @xmath168 ) * how much the symbols are dedicated to the training phase ? ( @xmath52 ) the problem ( p ) is obviously non - convex and is very complicated , but can be transformed into an equivalent problem .
since we deal with a static uplink scheduling policy , the achievable rate of user @xmath41 is independent to the sub - frame index @xmath43 and frame index @xmath44 if user @xmath41 is scheduled .
thus , we use the notation @xmath192 instead of @xmath193 $ ] if @xmath194 $ ] .
let @xmath195 and define the spectral efficiency of user @xmath41 ( in bps / hz ) as the average rate of user @xmath41 served by the bs normalized by time and bandwidth , given as @xmath196 } } } = \frac{1-{\frac{l}{n}}}{\eta } d_q r_j , \label{eq20 } \end{aligned}\ ] ] where @xmath197 denotes the bandwidth inefficiency ( such as the cyclic prefix overhead ) , and the second equality comes from ( [ eq18 ] ) .
since every user has the same throughput constraint @xmath31 , it is required that @xmath198 for @xmath199 , where @xmath200 is the common rate for scheduled group @xmath181 . inserting @xmath198 into ( [ eq20 ] ) ,
we have @xmath201 in order to guarantee the same minimum rate for all users , we further set @xmath202 from ( [ eq21 ] ) and ( [ eq22 ] ) , every user is guaranteed to exceed the common spectral efficiency @xmath203 our approach is first to maximize the spectral efficiency while providing the common spectral efficiency @xmath204 to every user in a cell at each possible value of @xmath52 . to meet the target throughput @xmath31 for every user
, the bs needs @xmath205 sub - frames so that the latency is given as @xmath206 where that last approximation is valid when @xmath207 is large .
then , an equivalent optimization problem can be formulated as follow : [ eq25 ] align ( ) & & & , + & & & r_j_q , j _ q , q , + & & & - .
note that a similar transformation is shown in @xcite . in the sequel
, we devise the optimal uplink scheduling policy by solving ( p - eq ) .
assume that @xmath52 is fixed and users @xmath209 are scheduled in the @xmath181th scheduling group and they are arranged in the descending order of @xmath210 , i.e. , @xmath211 , without loss of generality .
note that each scheduled user should have non - zero transmit energy for both the training and data transmission phases . in order to find the optimal transmit energy allocation in ( p - eq ) , the following sub - problem is considered .
[ eq26 ] @xmath212 to obtain the optimal solution , the following observations are helpful . * since @xmath192 is an increasing function of the transmit energy , the optimal transmit energy are obtained when @xmath213 for @xmath214 ( no waste ) , i.e. , @xmath215 * since @xmath216 for any @xmath217 , if @xmath218 is feasible , then @xmath219 is also feasible .
so , when @xmath10 users are scheduled , the optimal energy of the @xmath10th user should satisfy @xmath220 under the assumption that @xmath10 users are scheduled and the above two observations , the objective function of ( p - a ) in ( [ eq26]a ) can be written as @xmath221 from ( [ eq17 ] ) , where @xmath222 with @xmath223 and @xmath224 are defined as in table ii . since the objective function of ( p - a ) is a single - variable function , it can be easily solved and the following theorem states the optimal transmit energy allocation . for given @xmath52 and scheduling user set @xmath182 , the optimal transmit energy of user @xmath225 during the training and data transmission phases
is given as @xmath226 where @xmath227 is given in the bottom of the this page .
@xmath228 see appendix a. for later use , we define @xmath229 as the optimal common rate for given @xmath182 and @xmath52 .
note that by inserting ( [ eq31 ] ) into ( [ eq29 ] ) and using variables in table ii , it can be seen that @xmath230 is a non - decreasing function of @xmath231 and is independent to @xmath210 , @xmath232 .
even though the optimal transmit energy allocation strategy for given users are derived in ( [ eq30 ] ) , the size of the search space is too large to be exhaustively searched . to reduce the search space , we need to find implicit properties for the optimal scheduling groups . from ( [ eq23 ] ) , the objective function of ( p - eq ) is given by @xmath234 where @xmath235 is the sets of the optimal scheduling groups .
suppose that the cardinality of each of the optimal scheduling groups is given , i.e. , @xmath236 .
then , the problem ( p - eq ) is reduced to the following cardinality - constrained problem for each possible value of @xmath52 : [ eq33 ] @xmath237 to obtain the optimal scheduling groups , the following theorem is quite helpful .
denote @xmath238 , @xmath239 , as the optimal solution of ( p - b ) at a given value of @xmath52 .
then the optimal solution has the following properties : * for @xmath181 with @xmath240 , if @xmath241 , then there is no @xmath242 such that @xmath243 . * for @xmath181 with @xmath244 ,
if @xmath245 and there exists @xmath242 such that @xmath243 , then @xmath246 . see appendix b. from theorem 2 , it is shown that the optimal uplink scheduling policy is to select users having similar product values of the average allowed energy level @xmath247 and the path - loss @xmath248 and it significantly reduces the search space .
more detailed discussions on the search space will be given in sec .
iii - d .
although theorem 2 indicates some useful implicit properties for the optimal scheduling groups , it does not provide the exact solution explicitly , and we still need to find the optimal scheduling groups among the reduced search space . fortunately , it can be transformed into a binary integer programming ( bip ) with the following generic form : [ eq34 ] @xmath249 where @xmath250 $ ] is the @xmath251 state matrix , @xmath252 @xmath253 is the @xmath254 cost vector given by @xmath255^t},\end{aligned}\ ] ] @xmath256 is the @xmath254 all - one vector , @xmath257 denotes the number of candidate scheduling groups , and @xmath258 denotes the optimal common rate at given @xmath259 and @xmath52 , defined in ( [ eq32 ] ) .
the optimizing variable @xmath260 informs which candidate scheduling groups are selected , i.e. , if @xmath261 , the corresponding candidate scheduling group @xmath182 is selected as one of the optimal scheduling groups .
such a bip has been widely researched in literature and a variety of efficient algorithms are summarized in @xcite .
unfortunately , finding the optimal solution in a bip is known as np - hard in general . however , due to the special structure of our bip , it will be shown that a linear programming ( lp ) relaxation using @xmath262^{c\times 1}$ ] does not affect the optimality . to show this
, we introduce the following definition and two lemmas and then conclude the optimality of the proposed algorithm .
a matrix @xmath263 is _ totally unimodular _ if every square sub - matrix of @xmath263 has a determinant of @xmath264 , @xmath265 , or @xmath266 .
[ @xcite , example 7 ] if every column of a binary matrix @xmath263 has consecutive ones only without being interrupted by @xmath264s , then @xmath263 is totally unimodular .
[ @xcite , theorem 19.1 ] if @xmath267 is totally unimodular and @xmath268 is an integer vector , then the polytope described by @xmath269 , @xmath262^{c\times 1}$ ] , has integer vertices only .
now , we are ready to state the optimality of the proposed algorithm using the lp relaxation in ( [ eq34 ] ) .
the optimal solution of the bip in ( [ eq34 ] ) is identical to the solution obtained by using the lp relaxation on ( [ eq34 ] ) . from the properties of theorem 2 , every column of the matrix @xmath270 has consecutive ones only , which implies that @xmath270 is totally unimodular from lemma 4 .
using lemma 5 , the feasible region of ( [ eq34 ] ) is a polytope with integer vertices only , which guarantees that the solution obtained by using the lp relaxation on ( [ eq34 ] ) does not affect the optimality .
the proposed algorithm for obtaining the optimal static uplink scheduling policy is outlined in algorithm 1 .
the proposed optimal algorithm obtains @xmath271 and corresponding @xmath272 for each @xmath52 , and then find the optimal @xmath273 maximizing @xmath272 . the algorithm for obtaining @xmath271 is composed of the two parts .
the first part finds the candidate scheduling groups , denoted as @xmath274 , and their corresponding common rate by using ( [ eq32 ] ) obtained by using the optimal energy allocations in theorem 1 .
then , the second part finds the optimal combination of the selected scheduling groups that maximizes the spectral efficiency by applying the lp relaxation by virtue of theorems 2 and 3 .
_ example : _ here , we explain a toy example .
suppose that the network has @xmath275 users , @xmath276 , and @xmath277 and @xmath278kbits , @xmath279khz , @xmath280 , @xmath281 , @xmath282 , and @xmath283 .
the first part returns the candidate scheduling groups as @xmath284 and suppose that the corresponding common rate is determined as @xmath285 , respectively .
then , the cost vector and the state matrix are respectively set as @xmath286^t}\ ] ] and @xmath287.\ ] ] to satisfy the constraints ( [ eq34]b ) , there exist five feasible solutions : @xmath288^t$ ] , @xmath289^t$ ] , @xmath290^t$ ] , @xmath291,{\bf{x}}_5 = \left [ { 0,0,0,0,1,0,1 } \right]$ ] . since @xmath292 , @xmath293 and @xmath294 are selected as the optimal uplink scheduling groups . and @xmath295 and @xmath296 are the optimal scheduling portions and @xmath297 bps / hz .
then , the latency is @xmath298 frames .
note that among @xmath299 sub - frames @xmath300 sub - frames are allocated to @xmath301 and the remained sub - frames are allocated to @xmath302 .
define @xmath303 as the whole search space for finding the optimal scheduling groups in ( [ eq19 ] ) without theorem 2 , where @xmath304 is the collection of @xmath305-ary partitions of @xmath306 with at most @xmath1 elements , i.e. , each scheduling group size is no greater than the number of antennas , given by @xmath307 note that @xmath308 is well - defined only when @xmath309 and @xmath310 , where @xmath311 denotes the stirling number of the second kind . for a fixed @xmath305 , @xmath312 increases exponentially with @xmath33 .
thus , the whole search space is given as @xmath313 for a large @xmath33 .
now , define @xmath314 as the reduced search space for finding the optimal scheduling groups in ( [ eq19 ] ) with the aid of theorem 2 .
then , we can show that the cardinality of @xmath314 can be represented as the following recursive formula : @xmath315 which is known as the generalized fibonacci number @xcite .
with help of the binet s formula @xcite , we arrive at @xmath316 where @xmath317 is the unique positive root of @xmath318 . after some algebraic manipulations , @xmath319 and @xmath320 as @xmath138 so that @xmath321 , which implies that the reduced search space still increases exponentially with the number of total users , @xmath33 .
however , combined with theorem 3 , the following dramatical complexity reduction can be obtained . *
the reduction gain of theorem 2 itself also increases exponentially with @xmath33 .
in fact , the reduction gain is at least @xmath322 for a large @xmath33 . * without theorem 2 ,
the number of candidate scheduling groups in ( [ eq34 ] ) is @xmath323 , ( for @xmath324 , @xmath325 ) , which increases exponentially with @xmath33 .
however , due to theorem 2 , it reduces into @xmath326 , ( for @xmath324 , @xmath327 ) , which increases only squarely with @xmath33 . * as shown in theorem 3 , applying the lp relaxation on ( [ eq34 ] ) still provides the optimal solution of the original bip ( [ eq34 ] ) , which can dramatically reduce the exponential - time complexity into a polynomial - time complexity .
now , we are ready to quantify the computational complexity of algorithm 1 . the computational complexity of algorithm 1 consists of the following three parts , namely 1 ) the sorting operation ( line 1 ) , 2 ) the optimal energy allocation ( lines 3 - 9 ) and 3 ) solving the relaxed lp ( lines 10 - 15 ) . the worst - case computational complexity for sorting @xmath33 samples is @xmath328 .
since the optimal energy allocation requires @xmath329 iterations , the worst - case computational complexity of the second part is @xmath330 .
finally , the worst - case computational complexity of the lp is @xmath331 by using the karmarkar s algorithm @xcite .
thus , the total worst - case computational complexity for the proposed algorithm is @xmath332 .
in this section , we present some numerical results to verify the superiority of the proposed uplink scheduling policy .
one frame is set to occupy @xmath56mhz and @xmath266ms in the frequency and time domains and consists of @xmath333 sub - frames with @xmath334khz and @xmath266ms .
the number of symbols in each sub - frame is set to @xmath335 by assuming @xmath336 ( 25% cp overhead ) .
there are @xmath337 users each requesting @xmath338kbits date volume .
we use the pathloss model @xmath339 , where @xmath340 , @xmath341 and @xmath80 is given by @xmath342 with @xmath343 .
this pathloss model reflects the bs located at the origin and the users are located uniformly along the line @xmath344 $ ] .
all of users have the same transmit energy constraint , @xmath345 for @xmath168 . according to the simulation setting
, the received signal energy of the worst - case user at the bs is @xmath264db when @xmath346db energy is equally spread over the symbols in a sub-frame.dbm/hz for the noise spectral density , @xmath347db means only @xmath348dbm ( @xmath349mw ) per sub - frame in this simulation setting . ] when @xmath350 and @xmath351 .
the solid line represents the results using the zf receiver and the dashed line represents the results using the mrc receiver . ]
when @xmath352db and @xmath351 .
the solid line represents the results using the zf receiver and the dashed line represents the results using the mrc receiver . ]
when @xmath350 and @xmath351 . ]
when @xmath352db and @xmath351 . ]
the following three schemes are compared and simulation results using the zf receiver are summarized in tables iii - v : * ( random - equal ( re ) ) @xmath10 users are randomly selected and the transmit energy is equally used during the training and data transmission phases .
the training length , @xmath52 , and @xmath10 are exhaustively searched . *
( random - optimal ( ro ) ) @xmath10 users are randomly selected and the transmit energy is optimized by using theorem 1 . the training length , @xmath52 , and @xmath10 are exhaustively searched . * ( proposed ) the optimal uplink scheduling policy in algorithm 1 is used .
4 depicts the latency and spectral efficiency of the three schemes as a function of @xmath169 when @xmath350 and @xmath351 .
the solid line represents the results using the zf receiver and the dashed line represents the results using the mrc receiver .
when the zf receiver is employed , it is observed that at @xmath353db ( @xmath354dbm per sub - frame ) , the proposed uplink scheduling policy provides about @xmath355 or @xmath356 times smaller latency over the random - equal or the random - optimal scheme .
the major gain comes from the optimal energy allocation .
when the mrc receiver is employed , it is observed that at @xmath353db the proposed uplink scheduling policy provides about @xmath357 or @xmath358 times smaller latency over the random - equal or the random - optimal scheme .
similarly as in the zf case , the major gain comes from the optimal energy allocation .
the difference is that the common spectral efficiency of the zf receiver increases logarithmically with @xmath169 , while that using the mrc receiver is saturated at high @xmath169 due to the uncanceled interference .
so , when high @xmath169 is available , the zf receiver clearly outperforms the mrc receiver .
the gain of the zf receiver over the mrc receiver is @xmath359 and it becomes @xmath360 at @xmath353db .
5 depicts the latency and spectral efficiency of the three schemes as a function of @xmath1 when @xmath352db and @xmath351 .
when the zf receiver is employed , it is observed that at @xmath361 , the proposed uplink scheduling policy provides about @xmath362 or @xmath363 times smaller latency over the random - equal or the random - optimal scheme .
the gain of the proposed one over the random - optimal scheme diminishes when the number of bs antennas becomes high because all users can be scheduled with sharing the same resource .
similar trends can be observed when the mrc receiver is employed and the latency and common spectral efficiency of the two receivers become identical .
now , discussions on the behavior of the proposed uplink scheduling policy are provided .
6 depicts the optimal scheduling groups , @xmath364 $ ] , and the optimal training length , @xmath273 , as a function of @xmath169 when @xmath350 , @xmath351 , and the zf receiver is employed .
6 shows that small - size scheduling groups are preferred at low @xmath169 , while large - size ones are preferred at high @xmath169 , because high array gain is required at low @xmath169 . in spite of the pathloss difference ,
the size of each optimal scheduling group is nearly identical .
although a longer training period is required in high @xmath169 , it is less than or equal to @xmath365 , which is the half of each sub - frame . over a wide range of @xmath169
, @xmath273 is larger than or equal to @xmath366 for @xmath239 , which implies that orthogonal pilots can be used for not - so - large number of @xmath1 . also , small - size scheduling groups are preferred at low @xmath1 , while large - size ones are preferred at high @xmath1 because higher array gain is available .
7 illustrates the optimal scheduling groups , @xmath364 $ ] , and the optimal training length , @xmath273 , as a function of @xmath1 when @xmath352db , @xmath351 , and the zf receiver is employed .
similar trends are observed as in fig . 6 for the size of each optimal scheduling group . on the other hand ,
the optimal training length first increases with @xmath1 but it becomes decreasing if @xmath1 increases further .
interestingly , @xmath273 becomes smaller than @xmath366 for large @xmath1 , which implies that non - orthogonal pilots become beneficial .
in such a case , very high array gain is available and the optimal uplink scheduling policy provides an efficient non - orthogonal multiple access among users so that a low - latency ultra - reliable communication can be provided .
suppose that users are partitioned into @xmath367 for @xmath368 and figs .
8 ( a)-(c ) visualize the effect of the training length @xmath52 and the scheduling group size @xmath10 on the common spectral efficiency .
note that the support of @xmath369 is partitioned into two regions separated by the line @xmath370 , because non - orthogonal pilots are used in @xmath155 , while only orthogonal pilots are used in @xmath152 . in figs .
8 ( a ) and ( b ) , it is seen that the maximum points are belongs to the orthogonal region and the choice of the training length does not affect much for low @xmath169 and not high @xmath1 .
however , as can be seen in fig .
8 ( c ) , the optimal choice of @xmath52 does matter and it belongs to the non - orthogonal region at high @xmath1 .
so far , the superiority of the proposed uplink scheduling policy is verified by using some numerical examples , which indicates that the behavior of the optimal uplink scheduling policy changes significantly according to the system parameters @xmath169 and @xmath1 . in order to provided better insight on this , an asymptotic analysis would be fruitful .
it is interesting to consider the case that the total number of users , @xmath33 , and the number of bs antennas , @xmath1 are simultaneously large , but @xmath33 is far larger than @xmath1 .
the product of each user s location ( or pathloss @xmath371 ) and the transmit energy constraint @xmath372 is considered as independent random process with a common distribution @xmath373 and we denote @xmath374 $ ] as the averages of the transmit and the receive energies , respectively . the following theorem states the asymptotic behavior of the static uplink scheduling policy and its network latency .
let @xmath375 denote a random variable with cdf @xmath373 .
suppose that @xmath376 $ ] exists for @xmath377 given as in the bottom of this page .
then , as @xmath378 , the followings hold : * the asymptotically optimal pilot length , @xmath273 , and the optimal scheduling group size , @xmath379 , are respectively given by @xmath380 * and the network latency normalized by @xmath33 converges to @xmath381 where @xmath382 see appendix c. theorem 4 implies that a set of equi - sized scheduling groups become asymptotically optimal as @xmath378 .
so , it gives a hint to construct an easy way to implement an asymptotically optimal scheduling policy , which is outlined in algorithm 2 .
the computational complexity of the proposed asymptotically optimal scheduling policy is just @xmath383 , mainly comes form the sorting operation . since theorem 4 is involved with a complicated function @xmath377 , it is not easy to gain a good insight . for a further insight ,
we restrict the random variable @xmath210 with @xmath384 , which implies that almost all realizations of @xmath210 are scaled as @xmath0 . in addition
, the number of bs antennas is assumed to be large , but is much smaller than @xmath33 .
then , the following four scaling regimes according to the number of bs antennas , @xmath1 , and the average received energy constraint level @xmath0 , can be classified : * @xmath2 and @xmath385 : sufficiently high @xmath0 and not - so - large @xmath1 , * @xmath2 and @xmath386 : sufficiently high @xmath0 and large @xmath1 , * @xmath4 and @xmath387 : sufficiently low @xmath0 and not - so - large @xmath1 , * @xmath4 and @xmath388 : sufficiently low @xmath0 and large @xmath1 . then , the following theorem states the asymptotic behavior of the proposed optimal static uplink scheduling policy .
suppose that @xmath389 . as @xmath378 and @xmath137 with @xmath390 , the followings hold .
* the asymptotically optimal training length and the scheduling group size are respectively given by @xmath391 where @xmath108 is an arbitrary integer among @xmath392 , @xmath393 and @xmath394 denotes the lambert w function , known as the inverse function of @xmath395 @xcite .
* the asymptotically optimal network latency is given as @xmath396 see appendix d. from theorem 5 , some implications can be discussed as follows : * _ * regime i ) * _ : here , the transmit energy constraint is sufficiently high and the number of bs antennas is not sufficiently high .
so , this regime can be interpreted as the scenario that each bs equipped with a not - so - large number of antennas serves users with sufficient energy in a small - sized cell . in this case , the asymptotically optimal policy is to configure the half of each sub - frame as the training phase and to serve @xmath397 users simultaneously , which implies that orthogonal pilots are optimal .
note that this result is consistent to those in previous literature , such as theorem 2 in @xcite and sec .
v in @xcite .
also , equal - energy allocation over all symbols , i.e. , @xmath398 is nearly optimal , but the network latency can not be reduced as the number of bs antennas increases because the growth rate of the bs antennas is too slow . in case @xmath399 , the target throughput can be increased as @xmath400 while keeping the latency requirement , which is also consistent to the classical point of view . *
_ * regime ii ) * _ : the transmit energy constraint and the number of bs antennas are both sufficiently high .
so , this regime can be interpreted as the scenario that each bs equipped with a very large number of antennas serves users with sufficient energy in a small - sized cell . in this case , it is asymptotically optimal to configure one - third of a sub - frame as the training phase and to serve @xmath401 users simultaneously for each sub - frame , which implies that non - orthogonal pilots become optimal .
also , the optimal energy allocation is given as @xmath402 which implies that although @xmath403db higher energy per symbol is dedicated to each training symbol , but @xmath403db higher energy is allocated to the data transmission phase .
in this regime , the network latency can be arbitrarily reduced by increasing the number of bs antennas . however , allowing more energy is not beneficial . in case @xmath399 , the target throughput can be increased as @xmath404 while keeping the latency requirement or the latency is reduced as @xmath405 while keeping the target throughput .
* _ * regime iii ) * _ : here , the transmit energy constraint is quite tight and the number of bs antennas is not so high .
this regime can be interpreted as the scenario that each bs equipped with a not - so - large number of antennas serves users with limited energy in a large - sized cell . in this case , it turns out that optimal pilots are non - orthogonal and the optimal energy allocation is @xmath406 which implies that the energy allocation is identical to that in regime i ) , i.e. , equal - energy allocation over all symbols becomes optimal if @xmath407 is selected as in region i ) . in case @xmath399 , in order to meet the latency requirement , the target throughput needs to be scaled as @xmath408 . *
_ * regime iv ) * _ : the transmit energy constraint is quite tight , but the number of bs antennas is sufficiently high .
this regime can be interpreted as the scenario that each bs equipped with a very large number of antennas serves users with limited transmit energy in a large - sized cell . in this case , optimal pilots are non - orthogonal with the same asymptotically optimal scheduling policy to that for regime iii ) , but with different optimal energy allocation is given as @xmath409 however , if @xmath410 is selected as in regime i ) , the equal - energy allocation becomes optimal . in case @xmath399 , in order to meet the latency requirement , the target throughput needs to be scaled as @xmath411 .
it is worth noting that our analysis may be regarded as the results of the capacity - approaching receiver , even if we deal with simple linear receivers only because as @xmath412 , the lower - bound of the achievable rate of the zf or mrc receiver converges to the exact achievable rate , i.e. , @xmath413 and the achievable rate also converges to the capacity .
the only required condition is @xmath414 , which is valid in regimes i ) and ii ) .
however , this condition is not satisfied in regimes iii ) and iv ) , which implies that linear receivers become strictly sub - optimal if the network is operated in a limited energy regime . from the above
, it is shown that orthogonal pilots become optimal only in regime i ) , i.e. , only in a classical cellular system scenario , but non - orthogonal pilots become optimal in regimes ii)-iv ) , i.e. , in new scenarios for future cellular systems . to quantify the advantages of using non - orthogonal pilots , the asymptotically optimal network latency using orthogonal pilots only is given as follows .
suppose that @xmath415 but the network does not allow non - orthogonal pilots ( i.e. , @xmath416 ) . as @xmath378 and @xmath137 with @xmath390 , the asymptotically optimal network latency is given as @xmath417 since non - orthogonal pilots are not allowed , the asymptotically optimal network latency is obtained by using @xmath418 and @xmath419 .
as the number of bs antennas increases , the gain obtained by allowing non - orthogonal pilots becomes quite dramatic in regimes ii)-iv ) . in regimes
iii ) and iv ) , @xmath1 times lower network latency can be achieved . in regime
ii ) , the use of non - orthogonal pilots makes the network latency decrease sub - linearly rather than logarithmically with @xmath1 .
thus , an important design guideline can be derived : _ for a latency - sensitive application , it would be better to serve more users by employing non - orthogonal pilots , which is quite dramatic in the case of being operated in a high energy regime with a high target throughput , and is still quite meaningful even in the case of being operated in a low energy regime_. in order to prevent the misunderstanding on the above results , we would like to emphasize that each of users is assumed to have an independent transmit energy source for transmission in this paper .
so , as the number of scheduled users increases , more energy is consumed at a given sub - frame .
thus , in a low energy regime , i.e. , regime iii ) or iv ) , the gain coming from scheduling more users is larger than the channel estimation quality degradation , which is clearly different compared to the case in @xcite , where the total energy dedicated for a transmission is fixed regardless of the number of scheduled users .
in this paper , the latency - optimal static uplink scheduling policy is provided and its network latency is analyzed in an uplink training - based lsas employing simple zf or mrc receiver .
the optimal uplink scheduling problem considered in this paper is to minimize the network latency when each user is constrained with a target throughput and energy limit and the corresponding optimal solution provides the optimal scheduling groups with their own scheduling portions , the optimal energy allocation between the training and data phases for each user , and the optimal frame configuration for the training based lsas .
the optimal energy allocation is derived in a simple close - form for a given scheduling group and the optimal scheduling groups are found to be comprised of users with similar received signal quality .
then , a low - complexity uplink scheduling algorithm providing the exact optimal solution is proposed with polynomial - time complexity of @xmath420 . via numerical examples
, it is shown that the proposed uplink scheduling algorithm can provide an optimal policy which can provide several times lower network latency at given throughput and energy constraints over the conventional non - optimized scheduling algorithms in realistic environments , which shows that the proposed work can be a key enabler for the oncoming 5 g communication networks .
in addition , the proposed uplink scheduling policy and the corresponding optimal network latency are analyzed asymptotically to provide better insights on the system behavior . as the number of users increases , it is shown that the network latency , normalized by the number of total users , converges to a deterministic quantity and asymptotically optimal frame configuration and scheduling policy can be obtained , which gives a way to construct much simpler asymptotically optimal uplink scheduling policy with complexity of @xmath328 .
further , four operating regimes are classified according to the growth or decay rate of the average received signal quality and the number of bs antennas .
it turns out that orthogonal pilot sequences , widely used in current systems , are optimal only when the average received signal quality is sufficiently good and the number of bs antennas is not - so - large , i.e. , only in a conventional scenario . in other regimes representing new service and system scenarios , it turns out that non - orthogonal pilot sequences become optimal . by using non - orthogonal pilots ,
the network latency can be reduced by a factor of @xmath6 when the received signal quality is quite poor ( @xmath4 ) or by a factor of @xmath7 when the received signal quality is sufficiently good ( @xmath8 ) and the number of bs antennas is sufficiently large ( @xmath13 ) .
thus , this work proves that , in order to minimize the network latency , it is better to schedule users more than the amount of available training resource by employing non - orthogonal pilots for a training - based lsas , which would be a critical guideline for designing 5 g cellular communication systems supporting massive mtc ( or iot ) with low energy or latency - sensitive ultra - reliable tactile internet services .
for brevity , we consider the zf receiver only and drop the indices @xmath10 and @xmath52 . by taking the derivative of ( [ eq29 ] )
, we obtain @xmath421 since ( [ eq29 ] ) is non - negative and continuous on @xmath422 $ ] and @xmath423 , @xmath424 is obtained by finding a real root of the quadratic function @xmath425 as long as ( [ eq30 ] ) is feasible , i.e. , @xmath426 .
first , show that @xmath424 is real . to do this ,
it is sufficient that @xmath427 after inserting values in table ii , ( [ eq48 ] ) becomes equivalent to @xmath428 which is always true since @xmath429 .
now , show that @xmath430 . if @xmath431 , it is trivial so we omit .
suppose that @xmath432 .
then , the condition @xmath430 can be written as @xmath433 which is also always true since @xmath432 .
similarly @xmath430 when @xmath434 . finally , show that @xmath435 .
the case @xmath436 is again trivial so that we omit it .
assume that @xmath437 .
then , we have @xmath438 which holds since @xmath439 for any @xmath440 and inserting it in ( [ eq31 ] ) shows @xmath435 .
similarly for @xmath441 , we have @xmath442 which completes the proof .
without loss of generality , we assume @xmath443 and @xmath444 and rewrite the objective function as @xmath445 note that @xmath446 is a monotonically decreasing function of @xmath447 for @xmath448 and is independent to @xmath449 . since @xmath450 , in order to minimize @xmath451 , @xmath452 should be @xmath453 similarly , @xmath454 is successively determined once @xmath455 , @xmath456 , are determined , which concludes that the two properties in theorem 2 hold for all @xmath457 .
before proving theorem 4 , some preliminary results about a quantile function are introduced .
suppose that @xmath458 are i.i.d .
real - valued random variables with cdf @xmath34 and the order statistics of @xmath458 are denoted by @xmath459 . for @xmath460 ,
the @xmath461th quantile of @xmath34 is defined as @xmath462 .
correspondingly , the sample quantile is defined as the @xmath461th quantile of the empirical cdf @xmath463 with @xmath464 samples , @xmath465 , which can also be expressed as @xmath466 .
let @xmath467 be i.i.d .
random variables from a cdf @xmath34 satisfying @xmath468 for any @xmath469 . then , for every @xmath470 and @xmath471 , @xmath472 where @xmath473 and @xmath474 is a positive constant . the proof is directly obtained by applying the dvoretzky - kiefer - wolfowitz inequality @xcite . now , show the almost - sure convergence of @xmath475 as @xmath476 .
let @xmath467 be i.i.d .
random variables from a cdf @xmath34 .
then , @xmath477 . for
any given @xmath478 and sufficiently large @xmath35 , we have @xmath479 which can be made arbitrarily small by increasing @xmath35 because @xmath480 is convergent .
thus , @xmath481 as @xmath476 , which implies @xmath477 . for a riemann - integrable function @xmath482 ,
we have @xmath483 from lemma 7 , as @xmath476 , @xmath484 where the last convergence comes from the definition of the riemann integral . let @xmath485 be the @xmath305 samples of @xmath33 i.i.d random variables @xmath486 with @xmath487 .
define @xmath488 and assume that @xmath489 as @xmath378 .
then , @xmath490 since each @xmath491 and @xmath492 , @xmath493 as @xmath378 . from lemma 8 , we have @xmath494 since @xmath495 is continuous and each @xmath491 , small variation in the length of each interval ( @xmath496 vs. @xmath497 ) does not affect the convergence of the riemann integral so that @xmath498 which concludes the proof . in this proof
, we omit the index @xmath52 by assuming @xmath273 is used in the symbols @xmath499 , @xmath500 , @xmath501 , @xmath502 , and @xmath224 for simplicity .
assume that @xmath503 and let @xmath504 be the sequence of positive finite integers such that @xmath505 and @xmath506 , @xmath239 , where @xmath507 for @xmath508 .
we first prove that the optimal scheduling groups can be selected among equi - sized ones as @xmath378 . from ( [ eq23 ] ) and ( [ eq24 ] )
, we have @xmath509 where @xmath305 is the number of scheduling groups ( @xmath493 as @xmath378 ) and @xmath510 note that in ( [ eq29 ] ) , @xmath511 is given as @xmath512 since @xmath513 depends only on @xmath514 , denote @xmath515 by allowing some notational abuse . from lemma 9 as @xmath378 , @xmath516 which implies that the equi - sized scheduling group with size of @xmath379 can achieve the optimal network latency asymptotically . also from lemma 8
, the asymptotically optimal network latency can be expressed as @xmath517 first consider the case of @xmath518 and @xmath519 so that @xmath520 = 0 .
then , @xmath521 is given as @xmath522 inserting ( [ eq56 ] ) into ( [ eq54 ] ) , we obtain @xmath523 by using ( [ eq55 ] ) , we obtain @xmath524 which concludes the proof for the case .
the other cases can be shown in similar ways .
for @xmath525 , we obtain @xmath529 using the equality @xmath530 simplifies the case @xmath152 , @xmath531 , and inserting @xmath532 into ( [ eq61 ] ) results in @xmath533 by which we arrive at ( [ eq59 ] ) . on the other hand , by using taylor expansion , @xmath534 for @xmath527 .
then , for @xmath535 , we obtain @xmath536 to simplify the case @xmath537 , we use the identity @xmath538 by which we arrive at ( [ eq60 ] ) . by using the chebyshev s inequality ,
we obtain , as @xmath476 @xmath542 } \right| > c\mathbb{e}[{x_n } ] } \right ) \le \frac{{{\rm{var}}\left ( { { x_n } } \right)}}{{{c^2}{{\left ( { \mathbb{e}[{x_n } ] } \right)}^2 } } } \to 0,\ ] ] where @xmath543 is a finite and positive constant , which implies that the realizations of @xmath544 is included in the set @xmath545\le x_n\le(1+c)\mathbb{e}[x_n]\}$ ] almost - surely .
thus , the minimum and maximum in @xmath546 is scaled as @xmath547 $ ] almost - surely so that its transformation , i.e. , @xmath548 is also scaled as @xmath549)$ ] almost - surely by continuous mapping theorem @xcite , which completes the proof .
_ regime i ) _ : consider the case @xmath152 and let @xmath550 .
we obtain @xmath551 & = \theta \left ( { \frac{1}{{\log \left ( { 1 + a{\rho } } \right ) } } } \right)\\ & = \theta \left ( { \frac{1}{{\log { \rho } } } } \right),\label{eq63}\end{aligned}\ ] ] where the first equality comes from lemma 11 and the last equality comes from @xmath552 .
so , @xmath553 $ ] becomes independent to @xmath10 and @xmath52 asymptotically . by using theorem 5 and ( [ eq63 ] )
, we obtain @xmath554}{{k\left ( { 1 - \frac{l}{n } } \right ) } } \\&=\mathop { \arg \max } \limits_{k \le l } k\left ( { 1 - \frac{l}{n } } \right ) \nonumber \\&= \left(\frac{n}{2},\frac{n}{2}\right),\label{eq64}\end{aligned}\ ] ] and by inserting ( [ eq64 ] ) into ( [ eq43 ] ) , we obtain @xmath555\\ & = \theta \left ( \frac{\mathcal{t}_{\rm{th}}}{w{\log \rho } } \right),\label{eq65}\end{aligned}\ ] ] where the last equality comes from lemma 11 and @xmath556 .
now , consider the case @xmath557 . by using lemma 10 ,
we can rewrite ( [ eq41 ] ) as @xmath558 where @xmath559 if @xmath1 is finite , @xmath10 is also finite so that @xmath560 is maximized at @xmath561 and @xmath562 . inserting it into ( [ eq43 ] ) , we obtain @xmath563 comparing ( [ eq67 ] ) with ( [ eq65 ] ) indicates that the asymptotically lower network latency can be achieved when @xmath141 .
now , consider the case when @xmath564 .
for a fixed @xmath52 and @xmath564 , @xmath565 can be approximated as @xmath566 which is maximized at @xmath567 for some positive @xmath568 . ignoring the non - dominant term @xmath569 , inserting @xmath570 , and replacing @xmath52 with @xmath571 $ ] in ( [ eq68 ] ) yields @xmath572
by which we obtain optimal @xmath573 instead of using ( [ eq68 ] ) .
since ( [ eq69 ] ) is unimodal , it is sufficient to find the point @xmath574 satisfying @xmath575 and @xmath576 , which are respectively given as @xmath577 subtracting ( [ eq70 ] ) from ( [ eq71 ] ) results in @xmath578 and @xmath579 should satisfy @xmath580 for @xmath581 . after some mathematical manipulations , we arrive at @xmath582 from which we have @xmath583 and thus @xmath584 by introducing the lambert w function . by inserting @xmath585 in ( [ eq43 ] ) , we obtain @xmath586 combining ( [ eq65 ] ) and ( [ eq72 ] ) , we obtain @xmath587 .
_ regime ii ) _ : consider the case @xmath588 . then
, we obtain both ( [ eq65 ] ) and ( [ eq72 ] ) .
now , consider the case @xmath589 . for @xmath155 ,
we still obtain ( [ eq72 ] ) . however , for @xmath152 , @xmath590 so that @xmath591 = \theta(\frac{1}{\log m})$ ] .
although ( [ eq63 ] ) still holds , ( [ eq65 ] ) becomes @xmath592 , which , together with ( [ eq72 ] ) , results in @xmath593 .
_ regime iii ) and iv ) _ : from lemma 10 , ( [ eq41 ] ) can be rewritten as @xmath594}}{{k\left ( { 1 - \frac{l}{n } } \right)}}\\ & = \mathop{\arg\min}\limits_{{\scriptstyle 1\le l < n,\atop \scriptstyle 1\le k\le m } } \frac{{4n{{\log } _ 2}e}}{{k\left ( { m - k } \right)}}\mathbb{e}\left [ { { x^ { - 2 } } } \right],\label{eq73}\end{aligned}\ ] ] where the second equality comes from that @xmath595 for small @xmath20 .
finally , @xmath596 and @xmath273 can be selected arbitrarily among integers between @xmath266 and @xmath597 .
inserting @xmath598 , we obtain @xmath599 , which competes the proof
. 1 g. p. fettweis , `` the tactile internet : applications and challenges , '' _ ieee veh .
technol . mag .
_ , vol . 9 , no .
1 , pp . 6470 , mar .
t. taleb and a. kunz , `` machine type communications in 3gpp networks : potential , challenges , solutions , '' _ ieee commun . mag .
3 , pp . 178184 , mar .
l. atzori , a. iera , and g. morabito , `` the internet of things : a survey , '' _ comput .
15 , pp . 27872805 , oct .
j. c. ikuno , m. wrulick , and m. rupp , `` system - level simulation of lte networks , '' _ in proc .
ieee vtc spring _ ,
taipei , taiwan , may 2010 , pp . 15 .
g. piro , a. grieco , g. boggia , f. capozzi , and p. camarda , `` simulating lte cellular systems : an open source framework , '' _ ieee trans . veh .
498513 , feb .
2011 . t. l. marzetta , `` noncooperative cellular wireless with unlimited numbers of base station antennas , '' _ ieee trans .
wireless commun .
_ , vol . 9 , no . 11 , pp .
35903600 , nov .
f. rusek , d. persson , b. k. lau , e. g. larsson , t. l. marzetta , o. edfors , and f. tufvesson , `` scaling up mimo : opportunities and challenges with very large arrays , _ ieee signal process . mag .
_ , vol . 30 , no .
40 - 46 , jan . 2013 .
n. q. ngo , e. g. larsson , and t. l. marzetta , ' ' energy and spectral efficiency of very large multiuser mimo systems , `` _ ieee trans .
4 , pp . 14361449 , apr . 2012 .
a. muller , a. kammoun , e. bjornson , and m. debbah , ' ' linear precoding based on polynomial expansion : large - scale multi - cell mimo systems , `` _ ieee j. sel .
topics signal process . _ , vol . 8 , no .
5 , pp . 861875 , oct . 2014 .
l. zheng and d. tse , ' ' communication on the grassmann manifold : a geometric approach to the noncoherent multiple - antenna channel " , _ ieee trans .
inf . theory _ ,
2 , pp . 359383 , feb . 2002 .
b. hassibi and b. m. hochwald , `` how much training is needed in multiple - antenna wireless links ? '' _ ieee trans .
inf . theory _ , vol .
4 , pp . 951963 , apr . 2003 .
m. kobayashi , n. jindal , and g. caire , `` training and feedback optimization for multiuser mimo downlink , '' _ ieee trans .
59 , no . 8 , pp .
22282240 , aug .
s. murugesan , e. u .- biyikoglu , and p. schniter , `` optimization of training and scheduling in the non - coherent simo multiple access channel , '' _ ieee j. sel .
areas commun .
25 , no . 7 , pp .
1441456 , sep .
2007 . s. sun , and y. jing , `` channel training design in amplify - and - forward mimo relay networks , '' _ ieee trans .
wireless commun .
10 , pp . 33803391 , oct .
o. ayach , a. lozano , r. w. heath , `` on the overhead of interference alignment : training , feedback , and cooperation , '' _ ieee trans .
wireless commun .
_ , vol . 11 , no . 11 , pp .
41924203 , nov .
i. bettesh and s. shamai , `` optimal energy and rate control for minimal average delay : the single - user case , '' _ ieee trans .
inf . theory _
9 , pp . 41154141 , sep . 2006 . v. k. n. lau and c. ying , `` delay - optimal energy and subcarrier allocation for ofdma systems via stochastic approximation , '' _ ieee trans .
wireless commun .
_ , vol . 9 , no
. 1 , pp . 227233 , jan .
d. rajan , a. sabharwal , and b. aazhang , `` delay - bounded packet scheduling of bursty traffic over wireless channels , '' _ ieee trans .
inf . theory _ ,
1 , pp . 125144 , jan .
h. wang , w. zhang , y. liu , q. xu , and p. pan , `` on design of non - orthogonal pilot signals for a multi - cell massive mimo system , '' _ ieee wireless commun .
2 , pp . 129132 , apr .
shen , j. zhang , and k. b. letaief , `` downlink user capacity of massive mimo unser pilot contamination , '' _ ieee trans .
wireless commun .
14 , no . 6 , pp .
31833193 , jun . 2015 .
p. banelli , s. buzzi , g. colavolpe , a. modenini , f. rusek , and a. ugolini , `` modulation formats and waveforms for 5 g networks : who will be the heir of ofdm ? : an overview of alternative modulation schemes for improved spectral efficiency , '' _ ieee signal process . mag .
6 , pp . 8093 , nov 2014 .
b. m. hochwald , t. l. marzetta , and v. tarokh , `` multiple - antenna channel hargening and its implications for rate feedback and scheduling , '' _ ieee trans .
theory , _ vol .
9 , pp . 18931909 , sep .
2004 .
b. m. hochwald , t. l. marzetta , t. j. richardson , w. sweldens , and r. urbanke , `` systematic design of unitary space - time constellation , '' _ ieee trans .
inf . theory _ ,
19621973 , sep .
p. xia , s. zhou , and g. b. giannakis , `` achieving the welch bound with difference sets , '' _ ieee trans .
inf . theory _ ,
5 , pp . 19001907 , may 2005 .
d. calabuig , r. h. gohary , and h. yanikomeroglu , `` optimum transmission through the multiple - antenna gaussian multiple access channel , '' _ ieee trans .
inf . theory _ ,
1 , pp . 230243 , jan . | in this paper , an uplink scheduling policy problem to minimize the network latency , defined as the air - time to serve all of users with a quality - of - service ( qos ) , under an energy constraint is considered in a training - based large - scale antenna systems ( lsas ) employing a simple linear receiver .
an optimal algorithm providing the exact latency - optimal uplink scheduling policy is proposed with a polynomial - time complexity . via numerical simulations
, it is shown that the proposed scheduling policy can provide several times lower network latency over the conventional ones in realistic environments .
in addition , the proposed scheduling policy and its network latency are analyzed asymptotically to provide better insights on the system behavior .
four operating regimes are classified according to the average received signal quality , @xmath0 , and the number of bs antennas , @xmath1 .
it turns out that orthogonal pilots are optimal only in the regime @xmath2 and @xmath3 . in other regimes ( @xmath4 or @xmath5 ) , it turns out that non - orthogonal pilots become optimal .
more rigorously , the use of non - orthogonal pilots can reduce the network latency by a factor of @xmath6 when @xmath4 or by a factor of @xmath7 when @xmath8 and @xmath9 , which would be a critical guideline for designing 5 g future cellular systems . large - scale antenna system , training - based transmission , network latency minimization , uplink scheduling policy , non - orthogonal pilots |
hh 1 and 2 were the first herbig - haro ( hh ) objects to be discovered ( herbig 1951 ; haro 1952 ) , and have played an important role in the field of hh objects ( see the historical review of raga et al .
for example , hst images ( schwartz et al . 1993 ; hester et al . 1998 ) , proper motions ( ground based : herbig & jones 1981 ; hst : bally et al .
2002 ; ir : noriega - crespo et al .
1997 ; radio : rodrguez et al .
2000 ) , and detections in radio continuum ( pravdo et al .
1985 ) , uv ( ortolani & dodorico 1980 ) and x - rays ( pravdo et al . 2001 ) were first obtained for hh 1 and 2 .
the hh 1/2 system has a central source detected in radio continuum ( see , e.g. , rodrguez et al .
2000 ) and a bipolar jet system , with a nw jet ( directed towards hh 1 ) which is visible optically , and a se jet ( directed towards hh 2 ) visible only in the ir ( see noriega - crespo & raga 2012 ) .
hh 1 has a `` single bow shock '' morphology , and hh 2 is a collection of condensations , some of them also with bow - shaped morphologies ( see , e.g. , bally et al .
the emission - line structure of these objects has been studied spectroscopically , with 1d ( solf , bhm & raga 1988 ) and 2d ( solf et al .
1991 ; bhm & solf 1992 ) coverage of the objects .
it should be pointed out that the hh 1/2 outflow lies very close to the plane of the sky , so that projection effects do not have to be considered when interpreting the observations of these objects .
the spatial structure of the optical line emission has been studied at higher angular resolution with hst images .
schwartz et al .
( 1993 ) obtained h@xmath0 , [ s ii ] 6716 + 6730 and [ o i ] 6300 images .
later images of hh 1 and 2 were all taken with filters isolating the h@xmath0 and the red [ s ii ] lines ( bally et al .
2002 ; hartigan et al .
2011 ) . in the present paper
we describe a pair of new hst images of hh 1 and 2 obtained with filters isolating the h@xmath0 and h@xmath1 lines .
these images were obtained in consecutive exposures , so that they are not affected by proper motions ( which become evident in hst observations of the hh 1/2 complex separated by more than a few weeks ) nor by differences in the pointing , and they therefore yield an accurate depiction of the spatial distribution of the h@xmath0/h@xmath1 ratio in these objects .
these images show effects that have not been detected before in ground based studies of the emission line structure of hh 1 and 2 ( see , e.g. , solf et al .
1991 and bhm & solf 1992 ) nor in hst images of other hh objects ( since hst h@xmath1 images of hh objects have not been previously obtained ) . the paper is organized as follows .
the new hst images are described in section 2 .
the spatial distribution of the h@xmath0/h@xmath1 ratio , the line ratios as a function of h@xmath1 intensity and the distribution functions of the line ratios are presented in section 3 .
finally , an interpretation of the results is presented in section 4 .
the region around hh 1 and 2 was observed with the h@xmath0 ( f656n ) and h@xmath1 ( f487n ) filters on august 16 , 2014 with the wfc3 camera on the hst .
the h@xmath0 image was obtained with a 2686 s exposure and the h@xmath1 image with a slightly longer , 2798 s exposure .
the images were reduced with the standard pipeline , and a simple recognition / replacement algorithm was used to remove the cosmic rays .
the final images have @xmath6 pixels , with a pixel size of @xmath7 .
the images contain only two stars : the cohen - schwartz star ( cohen & schwartz 1979 ) and `` star no .
4 '' of strom et al . ( 1985 ) .
these two stars have been used to determine astrometric positions in ccd images of the hh 1/2 region since the work of raga et al .
( 1990 ) , yielding better positions for hh 1 ( which is closer to the two stars ) than for hh 2 .
we have carried out paraboloidal fits to the psfs of these two stars , and find no evidence for offsets and/or rotation , which shows the excellent tracking of the hst during the single pointing in which the two images were obtained .
also , we have analyzed the h@xmath8h@xmath1 difference images of the two stars , and find no offsets between the two frames .
the full h@xmath0 frame is shown in figure 1 , as well as blow - ups of regions around hh 1 and hh 2 in both h@xmath0 and h@xmath1 . as seen in the top frame
, the h@xmath0 map shows the extended collection of hh 2 knots ( to the se ) and the more compact distribution of the hh 1 knots ( towards the nw ) .
the central frames of figure 1 show the h@xmath0 emission of hh 2 ( left ) and hh 1 ( right ) at a larger scale . in the fainter h@xmath1 emission ( bottom frames of figure 1 )
only the brighter regions of hh 1 and 2 are detected .
we have defined two boxes ( labeled a and b in the bottom frame of figure 1 ) enclosing the regions of the two objects which are detected in h@xmath1 . in the following section ,
the regions within these two boxes are used in order to study the spatial dependence of the h@xmath0/h@xmath1 ratio .
frame of hh 1 and 2 obtained with the wfc3 camera of the hst ( the scale and orientation of the images is shown ) .
the central and the bottom frames show the h@xmath0 and h@xmath1 images ( respectively ) of regions containing hh 2 ( left ) and hh 1 ( right ) . also , on the h@xmath1 frames we show boxes which include the brighter regions of hh 1 and hh 2 ( boxes b and a , respectively ) , which have been used for calculating the h@xmath0/h@xmath1 ratios shown in figures 2 to 5 .
the images are displayed with a logarithmic greyscale.,width=302 ] as discussed in detail by odell et al .
( 2013 ) , the f656n filter is contaminated by emission from the [ n ii ] 6548 line , and both the f656n and f487n filters have contributions from the nebular continuum . using the fact that at all measured positions within hh 1 and 2 , the [ n ii ] 6548/h@xmath0 ratio does not exceed a value of @xmath9 ( see , e.g. , brugel , bhm & mannery 1981a and solf et al .
1988 ) and the transmission curve of the f656n filter ( see odell et al . 2013 and the wfc3 instrument handbook ) one then finds a peak contribution of @xmath10% to the measured flux . for estimating the effects of the continuum in the f656n and f487n images one can use the continuum and line fluxes obtained by brugel , bhm & mannery ( 1981a , b ) and the bandpasses of the filters to obtain estimates of @xmath11 and 5% ( for the f656n and f487n filters , respectively ) .
therefore , when interpreting the h@xmath0/h@xmath1 ratios obtained from our hst images , it is necessary to keep in mind that there is an uncertainty of @xmath12% due to a possible spatial dependence in the h@xmath1 line to continuum ratio within the f487n filter .
as this uncertainty is @xmath13 order of magnitude smaller than the h@xmath0/h@xmath1 ratio variations described below , we do not discuss it further .
figure 2 shows the h@xmath0 map ( right ) and h@xmath0/h@xmath1 ratio map ( left ) for hh 2 . to avoid having extended regions dominated by noise , in order to calculate the line ratio map it is necessary to place a lower bound on the line fluxes .
we have chosen to calculate the ratios only in regions in which the observed h@xmath1 flux is larger than @xmath14 erg s@xmath15pix@xmath15 . for calculating the intrinsic h@xmath0/h@xmath1 ratios we have applied the following reddening correction .
we first calculate the observed ratios for all of the pixels with h@xmath1 intensities larger than @xmath16 ( see above ) for the a and b boxes ( shown in the bottom frames of figure 1 ) . for hh 2
we obtain a mean line ratio @xmath17 , and for hh 1 an almost identical @xmath18 value . considering an observed line ratio of 3.8 for both objects , comparing with the case b recombination cascade intrinsic h@xmath0/h@xmath1 ratio of 2.8 and using the average galactic extinction curve
, we obtain an @xmath19 colour excess .
this value is somewhat lower than the @xmath20 value deduced for hh 2 by brugel et al .
( 1981a ) , using the method of miller ( 1968 ) , based on the fixed ratios between the auroral and transauroral lines of [ s ii ] ( i.e. , not assuming a recombination cascade h@xmath0/h@xmath1 ratio ) .
in order to calculate the dereddened h@xmath0/h@xmath1 ratios , we therefore multiply the observed ratios by a factor of 2.8/3.8 , basically assuming that the extinction towards hh 1 and 2 is position - independent .
the dereddened h@xmath0/h@xmath1 ratios of hh 2 ( see figure 2 ) have values in the @xmath21 range , with the regions of higher values corresponding to filamentary structures in the leading edge of the emitting knots ( i.e. , in the edges directed away from the outflow source ) .
in order to illustrate the positions of these `` high h@xmath0/h@xmath1 '' regions , we have superimposed an h@xmath0/h@xmath22 contour on the h@xmath0 emission map ( purple contour in the right frame of figure 2 ) .
figure 3 shows the h@xmath0 map ( bottom ) and dereddened h@xmath0/h@xmath1 ratio map ( top ) for hh 1 .
we have calculated the ratios only for pixels with an observed h@xmath1 flux larger than @xmath14 erg s@xmath15pix@xmath15 ( i.e. , the same cutoff used for hh 2 , see above ) .
the region with h@xmath0/h@xmath23 is a thin filament on the e side of the leading edge of hh 1 ( see the purple contour on the h@xmath0 emission map in the bottom frame of figure 3 ) .
it is clear that hh 1 shows a strong side - to - side asymmetry with respect to the outflow axis , as the sw region of the leading edge does not show high h@xmath0/h@xmath1 ratios ( see the top frame of figure 3 ) .
the h@xmath0 emission also shows a strong side - to - side asymmetry .
emission ( bottom , shown with a logarithmic colour scale ) and dereddened h@xmath0/h@xmath1 ratio ( top , shown with the linear colour scale given by the top bar ) for hh 1 .
the outflow source lies towards the sse . on the h@xmath0 image ( bottom frame )
we have included a ( dereddened ) h@xmath0/h@xmath1=4 contour ( in purple ) .
this contour shows that the higher values of the line ratio are distributed in a ridge along the e side of the leading edge of hh 1.,width=340 ] figure 4 shows the dereddened h@xmath0/h@xmath1 line ratio as a function of the ( observed ) h@xmath1 flux for all of the pixels with @xmath24 ( see above ) for hh 1 ( top frame ) and hh 2 ( bottom frame ) .
it is clear that for low values of the h@xmath1 intensity in both hh 1 and 2 we have a relatively broad distribution of line ratios ( the width of this distribution representing the relatively large errors of the line ratio at low intensities ) centered on the h@xmath0/h@xmath25 recombination cascade value . for pixels with brighter
intensities , we see a distribution of h@xmath0/h@xmath1 ratios extending from @xmath26 to larger values of @xmath12 ( for hh 1 ) or @xmath27 ( for hh 2 ) . /h@xmath1 ratio as a function of the observed h@xmath1 flux for the pixels of hh 1 ( top ) and hh 2 ( bottom ) . for the lower
h@xmath1 intensities we see that the line ratios straddle a value of @xmath28 ( shown with a horizontal line in both frames ) , corresponding to the recombination cascade h@xmath0/h@xmath1 ratio . for larger h@xmath1 intensities
we find line ratios extending from @xmath26 to @xmath29 ( for hh 1 ) or 6 ( for hh 2).,width=302 ] this result is seen more clearly in figure 5 , where we show the normalized distributions of the line ratios of pixels with @xmath30 erg s@xmath15 pix@xmath15 ( distribution @xmath31 , top frame ) , of pixels with @xmath32 erg s@xmath15 pix@xmath15 ( distribution @xmath33 , center ) , and of all pixels with @xmath34 ( distribution @xmath35 , bottom frame of figure 5 , with appropriate pixels found only in hh 2 ) . for both hh 2 ( left column ) and hh 1 ( right column of figure 5 ) , we see that the distribution @xmath31 of the lower intensity pixels is approximately symmetrical , centered at an h@xmath0/h@xmath36 line ratio .
the distributions for higher intensity pixels ( @xmath33 and @xmath35 , see above and the central and bottom frames of figure 5 ) start at values of h@xmath0/h@xmath37 - 2.5 , have a peak at a line ratio of @xmath38 and have a wing extending to h@xmath0/h@xmath39 for hh 1 and @xmath27 for hh 2 . in the following section ,
we show that these high h@xmath0/h@xmath1 ratios coincide with the values expected for collisional excitation of the @xmath4 and 4 levels of h. /h@xmath1 ratio bins for hh 2 ( left ) and hh 1 ( right ) .
we show three different distributions : @xmath31 ( top ) corresponding to pixels with @xmath14 erg s@xmath15 pix@xmath40 erg s@xmath15 pix@xmath15 , @xmath33 ( center ) of pixels with @xmath32 erg s@xmath15 pix@xmath15 and @xmath35 ( bottom ) of pixels with @xmath34 .
the distribution function of the lower intensity pixels ( @xmath31 , top frames ) is approximately symmetrical , centered at a line ratio of @xmath28 , corresponding to the recombination cascade value ( the dashed , vertical line in all plots corresponds to h@xmath0/h@xmath25 ) .
the distribution functions for higher intensity pixels ( @xmath33 and @xmath35 , central and bottom frames ) all show extended wings to higher values of h@xmath0/h@xmath1.,width=302 ]
from our new h@xmath0 and h@xmath1 hst images we can compute dereddened h@xmath0/h@xmath1 maps for the brighter regions of hh 1 and 2 .
for the reddening correction , we assume that the mean value of the h@xmath0/h@xmath1 ratio coincides with the recombination cascade value of 2.8 , as found previously by brugel et al .
( 1981a ) , who calculated the reddening correction with miller s method , based on the ratios of auroral to transauroral [ s ii ] lines .
we find that in limited spatial regions the ( dereddened ) h@xmath0/h@xmath1 ratio has values of @xmath41 , which are inconsistent with the recombination cascade value .
these high h@xmath0/h@xmath1 regions are filaments along the leading edges ( i.e. , the edges away from the outflow source ) of the brighter emitting regions of hh 1 and 2 ( see figures 2 and 3 ) . raga et al . ( 2014 ) show that the @xmath42h@xmath0/h@xmath1 ratio for a `` case b '' cascade fed by collisional excitations from the ground state of hydrogen has the approximate form : @xmath43 where @xmath44 is boltzmann s constant and @xmath45 is the energy difference between the @xmath46 and @xmath4 energy levels ( so that @xmath47 k ) .
the first term of this functional form has a temperature dependence derived from the ratio of the @xmath48 and @xmath49 collisional excitation coefficients ( assuming temperature - independent collision strengths ) , and the second term is a correction necessary to match the results of a 5-level , collisionally fed cascade matrix description of the hydrogen atom in the @xmath50k temperature range ( see raga et al . 2014 ) .
it is clear that the functional form of @xmath51 ( see equation [ r ] ) has high values for low temperatures , and has an asymptotic value of 4.35 for @xmath52 . from equation ( [ r ] )
, one obtains @xmath53 and @xmath54 .
therefore , the wing of the line ratio distributions extending to h@xmath0/h@xmath55 ( see figure 5 ) can straightforwardly be explained as produced in regions with temperatures in the @xmath56 k range emitting collisionally excited balmer lines .
this clear evidence that we are observing collisionally excited balmer lines together with the fact that the high h@xmath0/h@xmath1 regions are restricted to the leading edges of the outward moving condensations of hh 1 and 2 is quite conclusive evidence that we are observing the region of collisional excitation of h lines right after the shock waves driven into the surrounding medium by the condensations .
most of the h@xmath0 emission , however , comes from a region further away from the shock , in which the balmer lines are produced through the standard recombination cascade ( as evidenced by the h@xmath0/h@xmath57 ratios , see figures 2 and 3 ) . the theoretical prediction of these two regions of balmer line emission ( a collisionally excited balmer line region immediately after the shock , and the recombination region with balmer lines dominated by the recombination cascade ) in hh shock wave models
is already mentioned by raymond ( 1979 ) , and the h@xmath0 emission from the two regions was studied in more detail by raga & binette ( 1991 ) .
these two regions are of course present in all shock models ( for example , in the plane - parallel , time - dependent shock models of teileanu et al .
2009 ) . in non - radiative shocks observed in some supernovae remnants or in pulsar cometary nebulae , the observed
emission comes exclusively from the region of collisional excitation right behind the shocks ( see , e.g. , the review of heng 2010 ) . in hh objects , the only previous observational evidence of the emission from the immediate post - shock region ( as opposed to the emission from the recombination region )
were the h@xmath0 filaments seen in hst images of some bow shocks , notably in the hst images of hh 47 ( heathcote et al .
1996 ) , hh 111 ( reipurth et al . 1997 ) and hh 34 ( reipurth et al . 2002 ) .
however , as only h@xmath0 was observed it was not possible to guarantee that these filaments did correspond to the region of collisionally excited balmer lines .
our new h@xmath0 and h@xmath1 images for the first time show in a quite conclusive way that we have a detection of the immediate post - shock region of hh objects ( in which h is being collisionally ionized and the levels of h are being collisionally excited ) .
the detection of this region provides a clear way forward for developing models of hh bow shocks , in which the position of the shock wave relative to the recombination region is directly constrained by the observations .
we should note that throughout this paper we have assumed that the exctinction is uniform over the emission regions of hh 1 and 2 . in principle
it could be possible that foreground structures in the vicinity of the objects might produce changes in the extinction on angular scales comparable to the size of the objects .
however , estimates of the density of the pre - bow shock material of hh 1 and 2 ( based on observations of the post - shock density and on plane - parallel shock models , see , e.g. , hartigan et al . 1987 ) give values @xmath58 - 300 @xmath59 .
clearly , such a low density environment will not produce appreciable extinction on spatial scales comparable to the size of the hh objects .
because of this , if one wants to attribute the observed changes in the h@xmath0/h@xmath1 ratio to an angular dependence of the extinction , it is necessary to assume that still undetected , sharp - edged , high density regions are present in the immediate vicinity of hh 1 and 2 .
support for this work was provided by nasa through grant hst - go-13484 from the space telescope science institute .
ar and acr acknowledge support from the conacyt grants 101356 , 101975 and 167611 and the dgapa - unam grants in105312 and ig100214 . | we present new h@xmath0 and h@xmath1 images of the hh 1/2 system , and we find that the h@xmath0/h@xmath1 ratio has high values in ridges along the leading edges of the hh 1 bow shock and of the brighter condensations of hh 2 .
these ridges have h@xmath0/h@xmath2 , which is consistent with collisional excitation from the @xmath3 to the @xmath4 and 4 levels of hydrogen in a gas of temperatures @xmath5 k. this is therefore the first direct proof that the collisional excitation / ionization region of hydrogen right behind herbig - haro shock fronts is detected . |
particle current trough a coherent mesoscopic conductor connected at its left and right hand side to reservoirs is usually described in the non - interacting case by a formula , due to landauer , based on the following physical picture : electrons in the left ( right ) reservoir which are fermi distributed with chemical potential @xmath0 ( @xmath1 ) and inverse temperature @xmath2 ( @xmath3 ) can come close to the conductor and feed a scattering state that can transmit it to the right ( left ) reservoir . all possible dissipative processes such as thermalization occur in the reservoirs while the system formed by the conductor and the leads is assumed to be coherent .
the probability of being transmitted is a property of the conductor connected to the leads which is treated as a scattering system . in this picture , the probability that an outgoing electron comes back to the conductor before being thermalized is neglected , the contact is said to be reflectionless .
this description of the non - equilibrium steady state ( ness ) current through a finite system has been rigorously proved in some particular limiting situations @xcite , such as infinite reservoirs , but several difficulties prevail the understanding of non - equilibrium states in general and the description of current in more general situations , for instance in the case of interacting particles .
two frameworks are usually considered to study these open quantum systems : one deals with the properties of the state of the total ( infinite ) system @xcite where reservoirs are explicitly considered as part of the system .
the other is based on the master equation of the reduced density operator , obtained by tracing out the reservoirs degrees of freedom and is better suited to be applied to different systems and to compute explicitly some averaged ness properties , at the price of several approximations such as _ e.g. _ , born - markov ( see e.g. @xcite ) . in this paper
we explore particle current in a model where we mimic the leads that connect the reservoirs with the system , as a finite non - interacting system with a finite number of levels ( which we call meso - reservoir ) .
the reservoirs ( called here super - reservoirs ) are modeled by local lindblad operators representing the effect that a markovian macroscopic reservoirs have over the meso - reservoirs . in sec .
[ sec.2 ] we introduce the model and briefly review the method we use to solve it . in sec .
[ sec.3 ] we analyze the particle current operator and indicate the quantities that should be computed for a full description of the current . in sec .
[ landauer ] we briefly present the landauer formula that is expected to apply in some appropriate limits to our model and in sec . [ numeric ] we analyze the numerical results we obtained with our model and compare them with the current predicted by the landauer formula , validating the applicability of our model but also going beyond by computing the full probability distribution function ( pdf ) of the current . in sec .
[ conclu ] we present some conclusions and discuss interesting perspectives of our study .
we consider a one - dimensional quantum chain of spinless fermions coupled at its boundaries to meso - reservoirs comprising a finite number of spinless fermions with wave number @xmath4 ( @xmath5 ) .
the hamiltonian of the total system can be written as @xmath6 , where @xmath7 is the hamiltonian of the chain with @xmath8 the nearest neighbor hopping , @xmath9 the onsite potential and @xmath10 the annihilation / creation operator for the spinless fermions on the site @xmath11 of the chain ( conductor ) .
the chain interacts through the term @xmath12 with the meso - reservoirs @xmath13 . here
@xmath14 denotes the left and right meso - reservoir .
they share the same spectrum with a constant density of states @xmath15 in the band @xmath16 $ ] described by @xmath17 and @xmath18 are annihilation / creation operator of the left and right meso - reservoirs .
the system is coupled to the leads only at the extreme sites of the chain with coupling strength @xmath19 that we choose @xmath4-independent dependent super - reservoir to meso - reservoir couplings @xmath20 @xmath21 .
we assume that the density matrix of the chain - meso - reservoirs system evolves according to the many - body lindblad equation @xmath22 + \sum_{k,\alpha , m } \left ( 2l_{k , \alpha , m } \rho l_{k , \alpha , m}^\dag -\{l_{k , \alpha , m}^\dag l_{k , \alpha , m},\rho\ } \right ) , \label{eq : lindbladmodel}\end{aligned}\ ] ] where @xmath23 and @xmath24 are operators representing the coupling of the meso - reservoirs to the super - reservoirs , @xmath25 are fermi distributions , with inverse temperatures @xmath26 and chemical potentials @xmath27 , and @xmath28 $ ] and @xmath29 denote the commutator and anti - commutator , respectively .
the parameter @xmath30 determines the strength of the coupling to the super - reservoirs and to keep the model as simple as possible we take it constant .
the form of the lindblad dissipators is such that in the absence of coupling to the chain ( i.e. @xmath31 ) , when the meso - reservoir is only coupled to the super - reservoir , the former is in an equilibrium state described by fermi distribution .
@xcite .
to analyze our model we use the formalism developed in @xcite .
there it is shown that the spectrum of the evolution superoperator is given in terms of the eigenvalues @xmath32 ( so - called rapidities ) of a matrix @xmath33 which in our case is given by @xmath34 where @xmath35 and @xmath36 denote @xmath37 zero matrix and @xmath38 unit matrix , @xmath39 is the pauli matrix , and @xmath40 is a matrix which defines the quadratic form of the hamiltonian , as @xmath41 in terms of fermionic operators @xmath42 .
the ness average of a quadratic observable like @xmath43 is given @xcite in terms of the solution @xmath44 of the lyapunov equation @xmath45 with @xmath46 and @xmath47 as follows : consider the change of variables @xmath48 , the ness average of the quadratic observable @xmath49 is determined by the matrix @xmath44 through the relation @xmath50 .
wick s theorem can be used to obtain expectations of higher - order observables and in fact , the full probability distribution for these expectation values in some cases .
the operator representing the current flowing from the @xmath4-th level of the meso - reservoir to the chain is given by @xmath51 while the current trough the site @xmath52 of the chain is @xmath53 in the steady state the average current is conserved in this model @xcite and thus @xmath54 is independent of @xmath52 .
moreover , if we define the current from the left meso - reservoir as @xmath55 , we have that @xmath56 .
it is not difficult to note that current satisfies @xmath57 with @xmath58 now we are in a position to compute the full non equilibrium current distribution in terms of @xmath59 and @xmath60 . for this
we consider the generating function @xmath61 which , using eq.([moments ] ) , gives @xmath62 the probability distribution @xmath63 is the inverse fourier transform of @xmath64 , thus we get @xmath65 which is normalized .
we note that normality and positivity of probability lead an interesting inequality : @xmath66 an equivalent result holds for the current from the @xmath4-level of the meso - reservoir to the system : @xmath67 these expressions are expected because @xmath68 and @xmath69 are the possible eigenvalues of the operator @xmath70 ( similarly for @xmath71 ) , but they show that @xmath59 and @xmath60 contains all the information about the current .
we will study these quantities numerically , thus we need to solve the above mentioned lyapunov equation and note that in the @xmath72 variables ( we use the primed variables to indicate indices as they appear in variable @xmath73 , i.e. if @xmath11 is a site in the chain , then @xmath74 ) @xmath75 thus @xmath76 and @xmath77 wick s theorem implies @xmath78 to simplify the discussion , in what follows we reduce the number of parameters by assuming constant hopping and onsite energy @xmath79 and @xmath80 .
moreover we fix @xmath30 such that @xmath81 . in that case
, we showed @xcite that the transport quantities become roughly independent of @xmath30 .
in fact , in that case , the coupling super - reservoir meso - reservoir is stronger than that of the system ( chain ) meso - reservoir .
then the meso - reservoir are driven to a near equilibrium state weakly dependent of @xmath30 .
we explore now the behavior of the current as a function of @xmath15 , @xmath82 , @xmath83 and contrast the observation with expectations based on the landauer formula .
qualitative explanation of the current behavior is also provided .
the landauer formula @xcite provides an almost explicit expression for the ness average current as a function of the parameters of the system . in units
where @xmath84 and @xmath85 it reads @xmath86 where @xmath87\ , $ ] is the transmission probability written here in terms of @xmath88 the retarded and advanced green function of the system connected to the leads and of @xmath89 the imaginary part of the self - energy @xmath90 .
the self energies @xmath91 have only terms at the boundaries of the chains @xcite i.e. , @xmath92 , where @xmath93 if @xmath94 and @xmath95 if @xmath96 and @xmath97 recalling that both leads have the same spectrum , we assume a constant density of lead states @xmath98 in the range @xmath16 $ ] , thus @xmath99 is independent of @xmath100 inside the interval and zero otherwise . for the real part of the self - energy we have the principal value integral @xmath101
the landauer formula is expected to hold when the leads have a dense and wide spectrum .
therefore , we restrict ourselves to the case that @xmath102 and @xmath103 , the so called wide - band limit , where @xmath104 can be neglected .
the transmission coefficient is then @xmath105 and we need to compute the wide - band limit retarded green function @xmath106 note that in the previous expression we have set @xmath107 which sets the energy axis origin and @xmath108 which sets the energy scale .
thanks to a recursion relation , this matrix can be inverted @xcite and one explicitly finds the relevant element of the green function @xmath109^{-1}=\left(\omega + \frac{i \gamma _ l}{2}\right ) \left(\omega + \frac{i \gamma_r } { 2}\right ) \sum _ { k=0}^{\left\lfloor \frac{n-2}{2}\right\rfloor } ( -1)^k \omega ^{n-2 - 2 k } \binom{n-2-k}{k}\\ -\left(2 \omega + \frac{1}{2 } i ( \gamma_l + \gamma_r ) \right ) \sum _ { k=0}^{\left\lfloor \frac{n-3}{2}\right\rfloor } ( -1)^k \omega ^{n-3 - 2 k } \binom{n-3-k}{k}+\sum _ { k=0}^{\left\lfloor \frac{n-4}{2}\right\rfloor } ( -1)^k \omega ^{n-4 - 2 k } \binom{n-4-k}{k } , \end{array } \label{g1n}\ ] ] where @xmath110 is the largest integer smaller than @xmath111 . in the next section we compute numerically the integral in eq.([eq.landauer ] ) and compare with the results obtained in our model .
, width=264 ] in fig .
[ esquema ] we depict in blue and red two fermi distributions with @xmath112 levels and parameters @xmath113 and @xmath114 respectively . in the middle ( brown ) the spectrum of a chain with @xmath115 sites and @xmath116 the width of the chain energy band is @xmath117 with @xmath115 levels inside and is centered around @xmath107 . from this picture
, we expect that decreasing the width @xmath118 of the populated energy interval @xmath119 $ ] is equivalent to increase the width of the conduction band @xmath83 .
this is confirmed in the left panel of fig .
[ j - t ] where we show that the current is roughly independent of @xmath83 for @xmath120 and decreases with @xmath83 for @xmath121 ( black dots ) , when the conduction band extends beyond the region populated by electrons in the reservoirs .
the red dots are obtained for a larger @xmath118 for which the conduction band is always inside the populated region . in the rest of our numerical examples we set @xmath122 and @xmath123 .
analogously , in the right panel of fig .
[ j - t ] we show that for fixed @xmath83 the current grows linearly with @xmath118 and saturates at @xmath124 . for @xmath125 , @xmath126 , @xmath127 , @xmath128 and @xmath129 .
in the left panel ( black ) @xmath130 and ( red ) @xmath131 .
right panel : same parameters except : @xmath132.,title="fig:",width=264 ] for @xmath125 , @xmath126 , @xmath127 , @xmath128 and @xmath129 . in the left panel ( black )
@xmath130 and ( red ) @xmath131 .
right panel : same parameters except : @xmath132.,title="fig:",width=264 ] what is perhaps more interesting is the scaling of the current with @xmath133 . in fig .
[ j - vr ] we plot the current @xmath134 as a function of the coupling to the right lead , showing that the main trend of the current is @xmath135 . in the inset we show that there are deviations to this law . and @xmath136 .
the red dots are obtained by a numerical computation of the landauer formula with @xmath137 , and the continuous line is @xmath138 .
the inset is @xmath139 .
, width=264 ] we have analyzed this behavior using the landauer formula , that allows deeper analytical exploration .
since temperature is very low we take it exactly zero in both reservoirs , thus the landauer formula is @xmath140.$ ] changing the representation of the green matrix from position basis to the energy basis ( of the isolated chain ) one can see that it is possible to approximate ( isolated resonances approximation ) the integrand as a superposition of @xmath141 lorentzians each located at @xmath142 $ ] and with a width @xmath143/(n+ 1 ) , \
, j=1,\ldots n$ ] along the energy axis . if @xmath144 , the integral can be extended from @xmath145 to @xmath146 because the green function decays exponentially outside the energy band of the chain , thus the result @xmath147 is obtained .
now we can also explore fluctuations and compute @xmath148 . in fig.[fluct]a , we computed for the same parameters than in fig.[j - t]a the quantity @xmath149 .
we see that as soon as the current decreases because some modes of the chain goes out of the populated energy region , the fluctuation increases . as a function of other parameters like @xmath150 or @xmath141 ( data not shown ) ,
the average @xmath148 does not have important changes ( see fig.[fluct]b ) . as function of @xmath83 for @xmath151 ( black dots ) and @xmath152 ( red dots ) as in fig .
[ j - t]a .
panel b : @xmath149 for @xmath153 as a function of @xmath154 . in both figures
the other parameters are : @xmath155 and @xmath136 with @xmath156 in panel a and @xmath157 in panel b. , title="fig:",width=264 ] as function of @xmath83 for @xmath151 ( black dots ) and @xmath152 ( red dots ) as in fig .
[ j - t]a .
panel b : @xmath149 for @xmath153 as a function of @xmath154 . in both figures
the other parameters are : @xmath155 and @xmath136 with @xmath156 in panel a and @xmath157 in panel b. , title="fig:",width=264 ]
we showed that in the wide - band limit , the numerical results found in our model indeed correspond to what is expected on the basis of the landauer formula , a formula which is usually interpreted as if the reservoirs were always in an equilibrium ( gran canonical ) distribution , not perturbed by the presence of the system .
the landauer formula emphasizes the role of the fermi distributions of the reservoirs and provides an accurate description of the current if the assumption of reflectionless contacts is justified . in this respect ,
a very interesting relation was found in @xcite and proved in @xcite between the current and the occupation of the meso - reservoir @xmath158 $ ] .
this is an exact relation that in the appropriate limit should converge to the landauer formula .
note that it implies that the occupation difference with respect to the fermi distribution is @xmath159 .
it is a very interesting relation because it links the current , which is the fingerprint of the non equilibrium state , to the difference in distribution to the equilibrium case .
something similar has been found in classical systems where the fractal nature of the non equilibrium state is determined by the current @xcite .
moreover in @xcite we analyzed how onsager reciprocity relation is broken in the system and found that @xmath160 grows with @xmath30 implying that despite the almost @xmath30 independent value of the current , the dissipative mechanisms in the super - reservoir play an important role .
a deeper study of these effects , that are beyond the landauer picture , can be studied in the context of the model presented here and deserve further investigation .
the authors gratefully acknowledge the hospitality of nordita where part of this work has been done during their stay , within the framework of the nonequilibrium statistical mechanics program , and support from esf .
fb thanks fondecyt project 1110144 and sa thanks fondecyt project 3120254
. 99 w aschbacher , v jasi , y pautrat , and c pillet , j. math . phys .
* 48 * , 032101 ( 2007 ) .
s. ajisaka , f. barra , c.mejia-monasterio and t. prosen arxiv:1204.1321 s. ajisaka , f. barra , c.mejia-monasterio and t. prosen , arxiv:1205.1167 .
d. ruelle j. stat .
phys . * 98 * , 57 ( 2000 ) ; comm .
* 224 * , 3 ( 2001 ) .
s. ajisaka , s. tasaki and f. barra , bussei - kenkyu * 97 * 483 ( 2011/2012 ) ; arxiv:1110.6433 h. wichterich , m. henrich , h - p .
breuer , j. gemmer , and m. michel , phys .
e * 76 * , 31115 ( 2007 ) .
t. prosen , n. j. phys .
* 10 * , 3026 ( 2008 ) .
a. a. dzhioev and d. s. kosov , j. chem .
phys . 044121
* 134 * ( 2011 ) .
t. prosen j. stat .
p077020 ( 2010 ) .
s. datta , _ electronic transport in mesoscopic systems _
( cambridge university press , cambridge , 1995 ) . m. zilly , _ electronic conduction in linear quantum systems : coherent transport and the effects of decoherence _ phd .
dissertation universitt duisburg - essen ( http://duepublico.uni-duisburg-essen.de/servlets/derivateservlet/derivate-24279/dissertation_zilly.pdf ) .
f. barra and p. gaspard , j. phys . a math and gen . *
32 * 3357 ( 1999 ) .
p. gaspard , _ chaos , scattering and statistical mechanics _ ( cambridge university press , cambridge , 1998 ) . | we study particle current in a recently proposed model for coherent quantum transport . in this model a system connected to mesoscopic fermi reservoirs ( meso - reservoir ) is driven out of equilibrium by the action of super reservoirs thermalized to prescribed temperatures and chemical potentials by a simple dissipative mechanism described by the lindblad equation .
we compare exact ( numerical ) results with theoretical expectations based on the landauer formula . |
the recent upgrade of the very large array ( vla ) has resulted in a greatly increased imaging sensitivity due to the availability of large instantaneous bandwidths at the receivers and correlator .
at least two new dish array telescopes ( in particular , askap and meerkat ) are currently under construction to improve upon the vla s specifications in terms of instantaneous sky coverage and total collecting area .
a considerable amount of observing time has been allotted on all three instruments for large survey projects that need deep and sometimes high dynamic range imaging over fields of view that span one or more primary beams .
desired data products include images and high precision catalogs of source intensity , spectral index , polarized intensity and rotation measure , produced by largely automated imaging pipelines .
for these experiments , data sizes range from a few hundred gigabytes up to a few terabytes and contain a large number of frequency channels for one or more pointings . in this imaging regime ,
traditional algorithms have limits in the achievable dynamic range and accuracy with which weak sources are reconstructed .
narrow - band approximations of the sky brightness and instrumental effects result in sub - optimal continuum sensitivity and angular resolution .
narrow - field approximations that ignore the time- , frequency- , and polarization dependence of antenna primary beams prevent accurate reconstructions over fields of view larger than the inner part of the primary beam .
mosaics constructed by stitching together images reconstructed separately from each pointing often have a lower imaging fidelity than a joint reconstruction . despite these drawbacks , there are several science cases for which such accuracies will suffice .
further , all these methods are easy to apply using readily available and stable software and are therefore used regularly .
more recently - developed algorithms that address the above shortcomings also exist .
wide - field imaging algorithms @xcite include corrections for instrumental effects such as the w - term and antenna aperture illumination functions .
wide - band imaging algorithms such as multi - term multi - frequency - synthesis ( mt - mfs ) @xcite make use of the combined multi - frequency spatial frequency coverage while reconstructing both the sky intensity and spectrum at the same time .
wideband a - projection @xcite , a combination of the two methods mentioned above accounts for the frequency dependence of the sky separately from that of the instrument during wideband imaging .
algorithms for joint mosaic reconstruction @xcite add together data from multiple pointings either in the spatial - frequency or image domain and take advantage of the combined spatial - frequency coverage during deconvolution . such joint mosaic imaging along with a wideband sky model and wideband primary beam correction has recently been demonstrated to work accurately and is currently being commissioned @xcite(in prep ) .
these methods provide superior numerical results compared to traditional methods but they require all the data to be treated together during the reconstruction and need specialized software implementations that are optimized for the large amount of data transport and memory usage involved in each imaging run . with so many methods to choose from and various trade - offs between numerical accuracy , computational complexity and ease of use , it becomes important to identify the most appropriate approach for a given imaging goal and to quantify the errors that would occur if other methods are used .
the square kilometre array ( ska ) will involve much larger datasets than the vla , askap or meerkat will encounter with even more stringent accuracy requirements , making it all the more relevant to characterize all our algorithmic options and use existing , smaller instruments to derive and validate algorithmic parameters .
this paper describes some preliminary results based on a series of simulated tests of deep wide - band and wide - field mosaic observations with the vla .
section [ sec : sims ] describes how the datasets were simulated .
sections [ sec : algos : single1][sec : algos : mosaic ] list the imaging methods that were compared , for the single pointing as well as the mosaic tests .
section [ sec : metrics ] describes the metrics used to quantify imaging quality .
sections [ sec : results : single ] and [ sec : results : mosaic ] describe the results from several tests performed with the single - pointing and mosaic datasets .
section [ sec : discussion ] summarizes the results , discusses what one can and can not conclude from such tests , and lists several other tests that are required before ska - level algorithmic accuracy predictions can be made .
a sky model was chosen to contain a set of 8000 point sources spanning one square degree in area .
the source list is a subset of that available from the skads / scubed simulated sky project @xcite .
in this sample , intensities ranged between @xmath2 and @xmath3 and followed a realistic source count distribution . for high dynamic range tests , one @xmath4 source was also added .
spectral indices ranged between 0.0 and -0.8 with a peak in the spectral index distribution at -0.7 plus a roughly gaussian distribution around -0.3 with a width of 0.5 .
[ fig : scounts ] shows the source count vs intensity on the top - left panel and intensity vs spectral index on the bottom - left .
two types of datasets were simulated .
one was for a vla single pointing at c - config and l - band with 16 channels ( or spectral windows ) between 1 and 2 ghz .
the @xmath5-coverage was a series of snapshots the hpbw of the primary beam at l - band is 30arcmin and therefore covers the central part of the simulated region of sky .
the second dataset was for a vla mosaic at d - config and c - band with 46 pointings ( of primary beams 6arcmin in hpbw ) spaced 5 arcmin apart to cover roughly the same patch of sky at a comparable angular resolution . at these simulated angular resolutions ( 10arcsec at l - band c
- config and 9arcsec at c - band d - config ) , the expected confusion limit is @xmath6 , and the simulation included only sources brighter than @xmath7 in order to insulate these tests from errors due to main - lobe confusion .
16 channels ( or spectral windows ) were chosen to span the frequency range of 4 - 8 ghz , and the @xmath5-coverage corresponds to one pointing snapshot every 6 minutes , tracing the entire mosaic twice within 8.8 hours .
a true sky image cube was constructed by evaluating wideband point source components from the skads list for a set of frequencies that matched those being observed .
all sources were evaluated as delta functions ( naturally at pixel centers ) using the same cell size as would be later used during imaging .
specific tests with off - pixel - center sources were done by using different image cell sizes during simulation and imaging so that a source at a pixel center during simulation is not at the center during imaging .
visibilities were simulated per pointing for this image cube , using the wb - a - projection de - gridder @xcite which uses complex antenna aperture illumination functions to model primary beams that scale with frequency , have polarization squint and rotate with time ( due to the vla altitude - azimuth mount ) .
noise was not added to these simulations as our first goal was to characterize numerical limits purely due to the algorithms and their software implementations .
only after all observed trends and limits are understood will it be instructive to add gaussian random noise .
theoretically , pure gaussian random noise should not change the behaviour of algorithms other than increase error in predictable ways and it is important to systematically confirm that this is indeed the case in practice .
however , numerical noise will be present in these tests at the @xmath8 level as most image domain operations use single float precision .
all references to signal - to - noise ratio in this analysis therefore relate to numerical precision noise .
the datasets described above were imaged in a variety of ways . in all cases
the data products were continuum intensity images and spectral index maps .
the methods that were tested are several possible combinations of standard clean @xcite for narrow - band imaging , mt - mfs @xcite for wideband imaging , a - projection @xcite to account for direction - dependent effects during gridding , and stitched versus joint mosaics @xcite .
wideband primary beam correction was applied as appropriate using whatever primary beam models were available to the reconstruction algorithms .
all image reconstruction runs used the standard major and minor cycle iterative approach @xcite with different combinations of gridding algorithms for the major cycle ( prolate spheroid , a - projection ) and deconvolution algorithms for the minor cycle ( hogbom clean , mt - mfs ) .
the casa imaging software was used for these simulations and reconstructions as a combination of production tasks and custom c++ code and python scripts .
each frequency channel is reconstructed independently with standard narrow - band imaging algorithms ( clean ) .
there is no intrinsic primary beam correction , but deconvolution is followed by post - deconvolution primary - beam correction done per frequency .
all images are then smoothed to the angular resolution of the lowest frequency in the observation , spectral models are fitted per pixel to extract spectral indices , and channels are collapsed to form a continuum intensity image . the main advantage of this method is computational simplicity and ease of parallelization , with each channel and pointing being treated independently . the disadvantage of this approach is low angular resolution and possible sub - optimal imaging fidelity as the reconstruction process can not take advantage of the additional constraints that multi - frequency measurements provide .
this procedure is the same as cube , but with projection - based gridding algorithms applied per channel to account for baseline and time dependent primary beam effects .
aw - projection uses models of the antenna aperture illumination functions at different parallactic angles to compute gridding convolution functions per baseline and timestep .
these convolution kernels are constructed to have conjugate phase structure compared to what exists in the visibilities , and this eliminates beam squint during gridding .
the main expected differences from standard cube imaging is in the quality of the primary beam correction , visible in stokes v images all the time ( beam squint ) and at high dynamic range in the stokes i image .
multi - term multi - frequency synthesis was used to simultaneously solve for the sky intensity and spectrum , using the combined wideband @xmath5-coverage . with no intrinsic primary beam corrections ,
the output spectral taylor coefficients represent the time - averaged product of the sky and primary beam spectrum @xmath9 .
a wideband post - deconvolution correction , a polynomial division carried out in terms of taylor coefficients . ] of the average primary beam and its spectrum ( @xmath10 ) was done at the end to produce intensity and spectral index maps that represent only the sky .
this method has the advantage of algorithmic simplicity while taking advantage of the wideband @xmath5-coverage .
the main disadvantage is that the time variability of the antenna primary beam is ignored , which in the case of squinted and rotating beams can result in artifacts and errors in both the intensity and spectrum for sources near the half - power level .
also , since the frequency dependence of the primary beam persists through to the minor cycle modeling stage , the multi - term reconstruction has to model a spectrum that is steeper than just that of the sky .
more taylor terms are required , increasing cost and low snr instabilities .
multi - term multi - frequency synthesis is used along with wideband a - projection @xcite , an adaptation of aw - projection that uses convolution functions from conjugate frequencies to undo the frequency dependent effects of the aperture function during gridding in addition to accounting for beam rotation and squint .
this achieves a clear separation between frequency dependent sky and instrument parameters before the sky intensity and spectrum are modeled .
the output spectral taylor coefficients represent @xmath11 where the effective @xmath12 is no longer frequency dependent . in this case , a post - deconvolution division of only the intensity image by an average primary beam is required .
the output spectrum already represents only that of the sky brightness .
alternatively , a hybrid of cube imaging with narrow - band a - projection and mt - mfs can also be used in which the frequency dependence of the primary beam is removed in the image domain from the residual image cube before combining the frequency planes to form the taylor weighted averages needed for the mt - mfs minor cycle .
these two approaches have different trade - offs in numerical accuracy , computational load , memory use and ease of parallelization and a choice between them will depend on the particular imaging problem at hand .
this general approach has the advantage of accounting for the time and frequency dependence of the primary beam during gridding , and clearly separating sky parameters from instrumental ones .
the required number of taylor coefficients depend only on the spectrum of the sky , which is usually less steep than that of the primary beam . with a - projection , a flat - noise normalization choice can sometimes cause numerical instabilities around the nulls of the primary beam where the true sensitivity of the observations is also the lowest
work is in progress to find a robust solution to this .
alternate normalization choices will alleviate the problem but they will increase the degree of approximation that the minor cycle must now handle . in general , a mosaic can be constructed as a weighted average of single pointing images , using an average primary beam model as the weighting function . in this discussion , a combination after deconvolution will be called stitched mosaic , and a combination before deconvolution will be called a joint mosaic .
a joint mosaic can also combine the data in the visibility domain by applying appropriate phase gradients across the gridding convolution functions used by projection algorithms .
several wideband mosaic imaging options exist @xcite as various combinations of imaging with and without dd correction during imaging , cube versus multi - term multi - frequency synthesis imaging , and stitched versus joint mosaics . for cube clean imaging , joint mosaics
were made by combining data in the visibility domain and applying appropriate phase gradients across gridding convolution functions .
two methods were compared with the first using an azimuthally symmetric primary beam model to construct a single gridding convolution kernel for all visibilities and the second using full aw - projection to account for pb - rotation and beam squint ( per frequency ) .
the first method has the advantage of computational simplicity compared to full aw - projection where convolution functions can potentially be different for every visibility , but it has the disadvantage of ignoring beam squint and pb - rotation .
the primary beam model is also not the same as what was used to simulate the data and this test evaluates the effect of this commonly used simplifying assumption . for mfs imaging ( with multiple taylor terms )
, a joint mosaic was computed using aw - projection with its wideband adaptation ( to correct for the pb frequency dependence ) along with phase gradients applied to convolution kernels .
an alternate approach is to stitch together sets of pb - corrected output taylor coefficient images , using the time - averaged primary beam as a weighting function , and then recomputing spectral index over the mosaic .
however , initial tests showed that stitched mosaics ( with or without wb - awprojection ) produced larger errors than joint mosaics @xcite and stitched multi - term mfs mosaics were not included in this analysis .
the following metrics were used to evaluate the numerical performance of the different algorithms . 1 .
image rms : the rms of pixel amplitudes from off - source regions in the image , or in the case of no source - free regions , the width of a pixel amplitude histogram .
dynamic range : ratio of peak flux to peak artifact level or image rms when no artifacts are visible .
error distributions ( imaging fidelity ) : the intensity and spectral index maps produced by the above algorithms were compared with the known simulated sky and estimates of error per source were binned into histograms . for each output image ,
the simulated sky model image was first smoothed to match its angular resolution , and then pixel values were read off from both images at all the locations of the true source pixels .
histograms were plotted for @xmath13 where deviations from 1.0 indicate relative flux errors and for @xmath14 where deviations from 0.0 indicate relative errors in spectral index .
all histograms were made with multiple intensity ranges ( e.g.fig .
[ fig.lowdr.hist1 ] ) and over different fields of view ( e.g. fig .
[ fig.lowdr.hist2 ] ) to look for trends .
the l - band ( 1 - 2ghz ) c - configuration simulated data were imaged with the algorithms listed in sec . [
sec : algos : single1 ] through [ sec : algos : single4 ] . [ cols="^,^,^,^,^,^ " , ] +
the tests described in this paper address the use of wideband data for the deep imaging of crowded fields of compact sources at a sensitivity close to the confusion limit of the observation .
the simulations use a realistic source distribution ( from which only sources a few times brighter than the confusion limit were used ) and include primary beam effects arising from azimuthal asymmetry , parallactic angle rotation and frequency scaling .
imaging results were based on the ability to apply appropriate primary beam corrections and recover intensities and spectral indices of sources out to the 0.2 gain level of the primary beam and down to sensitivities a few times the confusion limit .
bright sources were introduced to evaluate dynamic range limits with and without corrections for the time variability of primary beams .
most observed trends were as expected and we were able to quantify the accuracy with which algorithms performed .
there were also a few surprises that highlight the unpredictability of current algorithms in certain situations . for a crowded field of compact sources being imaged at a sensitivity close to the confusion limit , the quality of the psf matters a great deal during deconvolution , and multi - frequency synthesis has a clear advantage over subband based imaging especially for weak sources .
the clean bias effect was seen for cube based methods which required careful masking to eliminate the effect .
however , mfs - based wideband algorithms had psfs with narrower main lobes and lower sidelobes and did not suffer from the clean bias thus making complicated masking procedures unnecessary . for a mosaic of such a field ,
a joint imaging approach is prefered ( within reasonable image size limits ) . for dynamic ranges higher than @xmath15 a - projection
based methods are required to account for baseline and time - dependent primary beam effects .
methods that derive spectral models of the sky while decoupling them from wideband instrumental effects ( mt - mfs with wideband a - projection ) are capable of producing usable spectral indices at all snrs except the lowest snrs ( where all methods fall short ) .
the use of wb a - projection even at dynamic ranges lower than @xmath15 might benefit imaging cases where strong sources exist in the outer parts of the pb where pb - spectral index is higher .
finally , simple data parallelization during the major cycle goes a long way in balancing out the increase in cost due to the more complex algorithms .
given the best possible approach ( multi - term mfs with wideband a - projection ) for the specific problem being studied ( deep widefield imaging of crowded fields at or near the confusion limit ) , we quantified errors in the recovered intensity and spectral index as a function of source snr for both single pointings and a joint mosaic .
note that for these noiseless simulations there still is numerical noise due to the use of single float precision in many image domain calculations . for our single pointing tests ( at l - band ) , errors in reconstructed intensity
were as follows .
sources brighter than @xmath16 have errors less than 5% , sources between @xmath16 and @xmath17 show errors at the 10% level and sources below @xmath17 show errors at the 20 to 30% level with several sources more than 50% .
errors in reconstructed spectral index were as follows .
sources brighter than @xmath16 show errors of @xmath18 , but sources between @xmath16 and @xmath17 show errors of @xmath19 . in these tests
, the weakest sources did not meet the threshold for spectral index calculation ( for both cube and multi - term mfs methods ) . for such crowded fields ,
accuracy also depends on the quality of the psf even if source snr is not a problem .
we showed that as psf sidelobe levels decreased ( by choosing different subsets of the data ) sources brighter than @xmath20 are always reconstructed to within a few percent , sources between 8 and 50 @xmath21 improve from 20% errors to less than 5% , and sources between @xmath22 and @xmath23 improve from over 50% errors to about 20% . for our joint mosaic tests
( at c - band ) , errors in reconstructed intensity were as follows .
sources brighter than @xmath24 had 2% errors , sources between @xmath24 and @xmath25 show errors of 4% and sources below @xmath17 have about @xmath26 errors .
spectral index errors were @xmath27 for sources brighter than @xmath16 but sources between @xmath16 and @xmath28 had errors upto @xmath29 with both cube and mfs methods showing a slight systematic bias of @xmath30 towards flatness .
the difference between the scale of the errors between the single pointing and mosaic tests relate to the amount of data used in the simulation .
there are many sources of error during image reconstruction and the systematic separation of various contributing factors and their eventual combination is crucial to building a complete picture and truly understanding the reasons behind observed effects .
conclusions derived from approximations done in isolation must be treated with caution and trends observed in real data must be reproduced when applicable . in a first step
we include artifacts typical of various instrumental effects ( primary beams , psf sidelobe levels ) and demonstrate clean bias and show how to eliminate it .
in fact , even the simulations done in this paper require the addition of several more effects to become accurate predictors of reality .
for example , this paper quantifies algorithmic limits in the situation of no noise , point sources located at pixel centers , and no differences between the primary beam models used for simulation versus those used during imaging ( except for one of the wideband mosaic tests ) .
these errors are thereforem a lower limit on what one could expect in reality .
effects such as the clean bias were reproduced clearly and its cause and solution understood .
future enhancements ( even just for stokes i imaging ) should include noise as well as residual calibration errors ( some of which may masquerade as effects needing baseline based calibration ) , the use of inaccurate primary beam models during imaging , the presence of extended emission in addition to point sources , etc .
such questions are listed in detail in sec .
[ sec : openqns ] . sometimes , algorithmic choices and achievable numerical accuracy depend on the type of available computing resources .
this section revisits the various algorithmic options in the context of numerical accuracy versus computational cost .
cube imaging methods are the easiest to parallelize , with both data and images being partitioned across frequency for the entire iterative imaging process .
there is minimal need for special - purpose software for such a setup .
imaging accuracy is limited to that offered by the @xmath5 coverage per channel , deconvolution depth is limited to the single channel sensitivity , and the resolution at which spectral structure can be calculated is limited to that of the lowest frequency .
there is no dependence on any particular spectral model which makes this approach very flexible in its reconstruction of spectral structure .
multi - frequency synthesis is demonstrably superior for continuum imaging due to its increased angular resolution , imaging sensitivity and fidelity , especially for crowded fields with thousands of compact sources .
the multi - term mfs algorithm is useful to compute in - band spectral indices along with intensity but the cost of both the major and minor cycle increase with the order of the polynomials used .
also , the accuracy of the spectral indices depends on the source snr and the choice of the order of the polynomial .
work is in progress to test an approach where the number of taylor terms is snr dependent . for multi - frequency synthesis ,
only the major cycle can be easily parallelized , with a gather step performed before the joint deconvolution step .
the prefered partition axis when the wb - awp algorithm is used is time because of the use of aperture illumination functions from conjugate frequencies during gridding . if narrow - band a - projection is used to form a cube before primary beam correction and the formation of taylor weighted averages in the image domain , the partition axis of choice for the major cycle would be frequency .
projection algorithms are significantly more expensive than the more standard method of using prolate spheroidal functions during gridding , mainly because of the support size of the convolution kernels and the overhead of computing such functions for potentially every visibility .
however , the more useful metric is the _ total _ runtime for imaging and the extra cost of using projection algorithms can be offset by a comparable reduction in the runtime due to its numerical advantages .
for example , wb a - projection increases the computing load for imaging but decreases the computing load and memory footprint of the mt - mfs setup which will need fewer terms to fit the spectral structure since the primary beam spectrum has been eliminated .
also , in practice the roughly 10 fold increase in computation due to the use of a - projection compared to the standard gridder is readily absorbed by simple data parallelization during the major cycle .
in addition , approximations can always be made ( identical antennas , coarser sampling of the aperture illumination function to reduce the size of the convolution functions , etc ) but effects of such approximations are visible beyond the @xmath31 dynamic range level .
another axis along which parallelization is relatively easy is pointing .
however , our tests show that the numerical differences between stitched vs joint mosaics are large enough that ( for crowded fields ) joint mosaics are always preferred .
another basic factor is the use of single vs double precision calculations during imaging and deconvolution .
currently ( in casa ) , all intermediate and output images use single precision , which is not the best option for dynamic ranges @xmath32 .
the simulations and tests described in this paper demonstrate several ways in which imaging accuracy can be sub - optimal even for the simple situation of point sources imaged using both traditional trusted techniques as well as newer ones .
a vast number of open questions and details remain , and a truly accurate picture can be derived only after these avenues are explored carefully and quantified to provide trends and usage guidelines to astronomers .
work is in progress on several of these fronts , and results will be presented in subsequent papers . 1 .
is it better to trade integration time at a single frequency band for shorter samples taken across a wider frequency range ?
for example for the vla , simulations have shown that comparable imaging sensitivities and far more accurate spectral indices are recovered when an observation spans multiple bands ( l - band and c - band for example ) compared with the entire time spent at only one band .
2 . what is the best algorithm for emission consisting of extended structure as well as compact emission
? algorithms like multiscale clean are usable but relatively more expensive in terms of computing and memory footprint load and require considerable human input .
several newer methods have been shown to produce superior results on their own but do not currently have production - quality optimized implementations that one can use .
3 . does the addition of noise and residual calibration errors change any of the above conclusions ?
theoretically , one would not expect the addition of gaussian random noise to change any results but it will be instructive to understand how robust these algorithms are to various noise levels .
residual calibration errors on the other hand might cause changes that must be quantified ( e.g. , see @xcite ) . to assess how well such effects can be corrected , traditional self - calibration techniques must be compared with more flexible methods such as direction dependent calibration schemes and peeling ( for example , sage ) , especially to test their effects on the accuracy of the reconstructed sources .
4 . how does baseline based averaging affect the achievable accuracy in the reconstructed intensity and spectral index ?
a popular mode of data compression is to average time - contiguous visibilities that will all fall on the same @xmath5 grid cell during gridding .
one concern with such an approach is whether it would prevent the handling of time variability of directional dependent instrumental effects or not . a simple test that achieved a data size reduction of 20% for one of the test datasets showed no noticeable effect with a - projection imaging out to twice the hpbw of the pb .
additional tests must be done with more practical data compression ratios .
how effective is the standard p(d ) analyses in predicting source counts below confusion limits ?
simulations similar to those described in this paper with sources weaker than @xmath7 ( or an observation with a larger angular resolution ) can be used to test the effect of main - lobe confusion for the simulated source count distribution and the accuracy of p(d ) analyses on such an image .
what type of software implementation and parallelization strategy is the most appropriate for a particular type of survey ? in the past few years
several new modern imagers have begun to become available and it would be instructive to repeat an imaging test with different algorithms and implementations to evaluate and quantify differences that arise simply from different software implementations and subtle numerical and algorithmic choices within it .
for example , @xcite shows examples of the numerical differences one can achieve simply by using pb models of different kinds and different algorithmic and software implementations for wideband mosaic imaging .
how do these results extend to full polarization imaging and faraday rotation synthesis ?
simulations with polarization dependent primary beams and full - stokes imaging ( with and without a - projection ) can quantify polarization imaging limits and identify appropriate imaging strategies .
work is in advanced stages @xcite .
to analyse this and produce the required algorithms and software to do such imaging .
how accurate do primary beam models need to be for use within a - projection ?
simulations with controlled differences in actual aperture illumination functions can give a useful idea of how much variation can be left unmodeled during imaging .
for example , it is easy to produce a spurious bias in spectral index simply by using a pb model whose shape is slightly different from what is present in the data .
work is in progress @xcite to quantity imaging errors for alma when ( not so ) subtle differences between antenna structures and illumination patterns are ignored at different levels of approximation and for the vla @xcite to carefully model primary beams from holography data and use them during image reconstruction .
these tests probe the limits of commonly used interferometric imaging algorithms in the context of crowded fields of compact sources being imaged at a few times the confusion limit and @xmath33 times the ( numerical ) noise level . in this regime ,
the quality of the psf is of considerable importance even at signal - to - noise ratios of @xmath34 simply because of the limitations of clean based deconvolution algorithms in crowded fields .
a psf sidelobe of @xmath35 to achieve errors of @xmath36 in intensity and @xmath37 in spectral index across a 1 - 2ghz band for low brightness sources near the confusion limit .
since psfs from the joint imaging of data ( mfs , for example ) typically have lower sidelobes compared to psfs from partitioned pieces of data ( cube , for example ) the former are prefered algorithmic choice in this regime . for shallow surveys where the emission above thermal noise limits fills the sky sparsely
, we did not find any statistically significant difference between algorithms that partition the data and those that do nt ( like mfs ) .
surveys that require imaging at the native resolution of the data / telescope particularly where detection and reconstruction of extended emission is important will still need to use mfs .
computing resources will be another discriminator in choosing algorithms in the shallow regime with algorithms that do nt require data partitioning ( like mfs ) requiring fewer resources than algorithms that do require partitioning . for wide - field imaging ( dr @xmath38 ) ,
primary beam correction methods need to include details such as its time , frequency and polarization dependence to both reconstruct the bright source accurately and to eliminate artifacts that may contaminate surrounding weaker sources .
our investigation shows that accounting for azimuthal asymmetry of the beams , rotation or pointing jitter as a function of time , scaling with frequency and polarization beam squint due to off - axis feed locations is certainly required .
the imaging performance of joint image reconstruction methods is fundamentally superior to those that work with partitioned data and combining number of reconstructed images each from a fraction of the available data .
hybrid implementations that take advantage of the ease of parallelization of partitioned methods where possible may be useful , but require careful analysis of its final imaging performance .
work presented here also gives a lower limit on the level of detail at which simulations for future surveys and telescopes must be undertaken .
it is important to systematically build up the complexity of the simulation in a way that enables one to efficiently pinpoint the reasons behind observed feature or trends . given the wide range of observation types and analysis methods , simulations must be specific to the parameters of each survey and must be as complete as possible in their inclusion of instrumental effects and observing modes .
in particular , effects of time , frequency and polarization dependence of the antenna primary beams , effects of long baselines and wide fractional bandwidths , and correct sampling of the coherence field in time and frequency to realistically reflect both the data volume and the filling factor in the uv - plane .
clear accuracy requirements are also necessary for each survey in order to choose the optimal analysis procedure based on scientific estimates .
we wish to thank the various nrao staff members for useful discussions at various stages of this project .
we wish to thank the common astronomy software applications ( casa ) group for the use of their imaging libraries in our simulation and imaging scripts . | many deep wide - band wide - field radio interferometric surveys are being designed to accurately measure intensities , spectral indices and polarization properties of faint source populations . in this paper
we compare various wideband imaging methods to evaluate the accuracy to which intensities and spectral indices of sources close to the confusion limit can be reconstructed .
we simulated a wideband single - pointing ( c - array , l - band ( 1 - 2ghz ) ) and 46-pointing mosaic ( d - array , c - band ( 4 - 8ghz ) ) jvla observation using realistic brightness distribution ranging from @xmath0jy to @xmath1jy and time- , frequency- , polarization- and direction - dependent instrumental effects .
the main results from these comparisons are ( a ) errors in the reconstructed intensities and spectral indices are larger for weaker sources even in the absence of simulated noise , ( b ) errors are systematically lower for joint reconstruction methods ( such as mt - mfs ) along with a - projection for accurate primary beam correction , and ( c ) use of mt - mfs for image reconstruction eliminates clean - bias ( which is present otherwise ) .
auxiliary tests include solutions for deficiencies of data partitioning methods ( e.g. the use of masks to remove clean bias and hybrid methods to remove sidelobes from sources left undeconvolved ) , the effect of sources not at pixel centers and the consequences of various other numerical approximations within software implementations .
this paper also demonstrates the level of detail at which such simulations must be done in order to reflect reality , enable one to systematically identify specific reasons for every trend that is observed and to estimate scientifically defensible imaging performance metrics and the associated computational complexity of the algorithms / analysis procedures . |
we are interested in rg flows in two - dimensional euclidean quantum field theories .
we will look at flows that originate in a uv fixed point described by a 2d @xmath0 and arrive at an ir fixed point described by a @xmath1 which may be trivial if a mass gap develops along the flow .
the flows are triggered by perturbations of @xmath0 by relevant operators @xmath2 .
in general it is a hard non - perturbative problem to determine the infrared properties of the perturbed theory , in particular to identify @xmath1 when it is non - trivial .
a technique that allows one to investigate the perturbed theories numerically in the infrared including the flows to non - trivial fixed points is the truncated conformal space approach ( tcsa ) invented in @xcite .
the basic setup of tcsa is as follows . in order not to worry about perturbative infrared divergences we put the perturbed theory on a cylinder with spacial periodic coordinate @xmath3 and euclidean time @xmath4 directed along the axis of the cylinder . in the hamiltonian formalism
the perturbed hamiltonian on a circle at @xmath5 is [ ham_pert ] h = h_0 + ^i dx _ i(x,0 ) . here
( l_0 + |l_0 - ) is the hamiltonian of @xmath0 that is expressed via the virasoro modes @xmath6 , @xmath7 and the central charge @xmath8 .
the eigenvalues of @xmath9 are scaling dimensions of @xmath0 shifted by @xmath10 .
since the state space @xmath11 of the unperturbed theory on a cylinder is discrete @xmath12 can be represented by an infinite matrix acting in this space . to regulate the uv divergences we can truncate @xmath12 to a finite matrix by restricting it to a finite dimensional subspace in @xmath11 .
this gives rise to a variety of truncated hamiltonian techniques in which one numerically calculates the eigenvalues and eigenvectors of the truncated hamiltonian matrix for various values of the scale set by the circle length @xmath13 . in tcsa
the truncated subspace is specified by imposing an upper bound on the scaling dimensions of the allowed states .
this upper bound is called the truncation level .
the uv divergences show up in the dependence of numerics on the truncation level .
this dependence and various improvement techniques have been discussed in @xcite , @xcite , @xcite , @xcite .
tcsa has been applied to situations in which a perturbed theory arrives to a non - trivial fixed point ( see e.g. @xcite , @xcite ) . in this case the dimensionless energy eigenvalues @xmath14 at large @xmath13 asymptote to constant values that give scaling dimensions in @xmath1 .
moreover the asymptotic eigenvectors that correspond to scaling states @xmath15 in @xmath1 are obtained as vectors in the truncated subspace of @xmath11 .
the corresponding components of @xmath16 give us pairings [ tcsa_pairing ] i | v_j between scaling states in both cft s .
pairings of this type are also associated with conformal interfaces .
a conformal interface @xmath17 between @xmath0 and @xmath1 can be described ( via the folding trick @xcite ) as a conformal boundary condition in the tensor product @xmath18 .
if @xmath19 and @xmath20 are scaling fields in the two theories then we can define a pairing [ conf_pairing ] _ uvi |j_ir = _ i^uv _ j^ir _ i as a disc one - point function with boundary condition @xmath17 .
such pairings are canonically normalised by means of cardy constraint @xcite . in tcsa
there is no canonical way to fix normalisation of the eigenstates @xmath21 so the natural observables that are free from this ambiguity are component ratios [ gen_rat ] ^j_i , k = . in ( untruncated )
quantum field theory perturbed states more often than not do not lie in the unperturbed state space .
this has many manifestations such as haag s theorem , orthogonality catastrophe , inequivalent representations of canonical commutation relations , etc ( see @xcite for a nice exposition and discussion ) .
typically however the problem is with the norm of the perturbed states in interaction representation which is formally infinite , while the component ratios such as ( [ gen_rat ] ) are well defined . to show that the pairing ( [ tcsa_pairing ] ) and ratios ( [ gen_rat ] ) associated with an rg flow arise from a particular local conformal interface we discuss states in lagrangian formalism using wave functionals .
suppose @xmath0 is described via a fundamental field @xmath22 and an action functional @xmath23 $ ] .
states in this theory can be represented by a wave functional @xmath24 $ ] depending on functions @xmath25 defined on the circle @xmath5 . the vacuum state is then given by a renormalised functional integral over the left half cylinder @xmath26 _ vac[_0 ] = _ ( x,0)=_0(x ) e^-s_0 [ ] .
the functionals describing excited states can be obtained by inserting local operators at positions with @xmath27 inside the above functional integral .
suppose now that the perturbed theory is described by an action functional s [ ] = s_0 [ ] + dxdy v((x , y ) ) where @xmath28 is some potential .
the vacuum of the perturbed theory can be represented by the functional integration over the fields defined [ perturbed_psi ] _
vac^pert[_0 ] = _ ( x,0)=_0(x ) e^-s [ ] .
expanding inside the functional integral the exponent @xmath29\ ] ] in power series we obtain a formal expansion of the perturbed vacuum functional @xmath30 $ ] in terms of unperturbed states ( interaction representation ) .
we can extend this pairing to excited states of the deformed theory by inserting into the functional integral ( [ perturbed_psi ] ) additional local operators .
this gives a pairing between the states in the two theories .
in particular if we follow the perturbed theory to the ir fixed point we obtain a pairing of the type ( [ tcsa_pairing ] ) .
we assume in this discussion that all divergences are renormalised including the ones that need additional boundary counterterms at @xmath5 .
renormalisation of wave functionals has been discussed in @xcite , @xcite , @xcite and more recently in @xcite .
( more on renormalisation shortly . )
what becomes clear in this picture is that the pairing arises by perturbing the uv fixed point on a half space ( a half cylinder ) and letting the perturbed theory flow to the ir fixed point .
this gives us a local conformal interface between @xmath0 and @xmath1 which we call an rg interface .
the idea to associate such interfaces with rg flows was put forward in @xcite ( see also @xcite ) .
a concrete proposal for such an interface for @xmath31-flows between neighbouring minimal models was put forward in @xcite and pairing ( [ conf_pairing ] ) associated with it was considered . note that we do not need to have fundamental fields and functional integral representation to define the pairing in terms of the rg interface . to elucidate renormalisation and to relate the above functional integral picture to conformal perturbation and tcsa we first recast ( [ perturbed_psi ] ) in the language of conformal perturbation .
wave functionals correspond to a particular basis in state space in which the field operator @xmath32 is diagonal .
we can choose instead the conformal basis of scaling states @xmath33 in @xmath0 .
consider the following amplitude on a cylinder of length @xmath34 [ z0i ] z_0,i(l ) = 0| | i illustrated on the picture below .
( -2,0 ) ( 6,0 ) arc ( 270:90 : 0.6 and 1.1) ( -2,2.2 ) arc ( 90:270 : 0.6 and 1.1)cycle ; ( -2.25,0)(6.25,0 ) ; ( -2.25,2.2)(6.25,2.2 ) ; ( -2.25,2.2 ) arc ( 90 : 270 : 0.6 and 1.1 ) ; ( -2.3,0 ) arc ( -90:90:0.6 and 1.1 ) ; ( -2,2.2 ) arc ( 90 : 270 : 0.6 and 1.1 ) ; ( -2,0 ) arc ( -90:90:0.6 and 1.1 ) ; ( 6,0 ) arc ( 270:90:0.6 and 1.1 ) ; ( 6.25,1.1 ) circle [ x radius=0.6 , y radius = 1.1 ] ; ( 6.0,0 ) arc ( -90:90:0.6 and 1.1 ) ; ( 7.3,1.1 ) node @xmath35 ; ( -3.3,1.1 ) node @xmath36 ; ( 2.3,1.1 ) node @xmath37 ; ( 2.2 , 0.6 ) node perturbed theory ; ( -2,2.25 )
( -2 , 2.6 ) ; ( 6,2.25)(6,2.6 ) ; ( -2 , 2.45 )
( 6,2.45 ) ; ( 2,2.68 ) node @xmath34 ; we have added little collars
( depicted white ) of unperturbed theory at both ends where the states @xmath36 , @xmath33 are attached .
the actual width of the collars is inessential .
the perturbative expansion of ( [ z0i ] ) is z_0,i(l ) = _ n _ i_1 ,
, i_n _ -ly_10 d^2 z_1 _ -ly_n0 d^2 z_n0| _ i_1(z_1 )
_ i_n(z_n ) |i the correlators here are correlators in @xmath0 and can be calculated by mapping the cylinder onto an annulus on the plane and inserting @xmath2 at infinity .
divergences arise when operators @xmath38 collide .
collisions can happen at a point in the bulk of the cylinder ( annulus ) or at a point on the boundary : at @xmath5 or @xmath39 .
the former are handled by the usual renormalisation of perturbed theory while the latter may give rise to additional boundary counter terms of the type discussed in @xcite ( and @xmath30 $ ] ) ] .
it is clear from the collisions picture that all divergences are local .
assuming they can be renormalised we obtain a cylinder with two local interfaces between @xmath0 and the perturbed theory with two external states of @xmath0 attached .
the perturbed theory can be driven towards @xmath1 that results in having a system with two conformal interfaces . taking the limit @xmath40 results in the following asymptotic z_0,i(l )
~e^-le_0^ir 0|v_0v_0|i where @xmath41 is the vacuum energy of @xmath1 .
removing the divergent exponential factor we obtain up to the overall factor @xmath42 the overlaps @xmath43 of the perturbed vacuum @xmath44 with the unperturbed scaling states .
overlaps with excited states @xmath15 can be obtained from the subleading terms in the @xmath40 asymptotic . to summarise the above discussion , in the lagrangian formalism the tcsa overlaps ( [ tcsa_pairing ] )
are described in terms of a local interface between the perturbed and unperturbed theories .
this has implications that are not evident in the hamiltonian picture ( and thus in tcsa ) .
thus for flows to a trivial fixed point the rg interface is a conformal boundary condition ( the rg boundary ) and must satisfy cardy constraint which arises from locality . at the ir fixed point
the perturbed theory vacuum state is given by a conformal boundary state which is a particular combination of ishibashi states satisfying cardy constraint .
there are similar constraints for flows to a non - trivial fixed point with local conformal interfaces described by conformal boundary conditions in the tensor product theories as was already mentioned above .
all massive flows from a given @xmath0 are then labeled by conformal boundary conditions .
they appear to be natural labels of infrared phases of massive theories .
the unstable manifold is then broken into regions of possibly different dimensions labeled by conformal boundary conditions with particular conditions on boundaries separating these regions .
it seems to be an interesting enterprise to investigate this structure for particular two - dimensional cfts . for each virasoro minimal model
there is a finite number of fundamental conformal boundary conditions , so that we expect their unstable manifolds to be broken into finitely many regions ( massive phases ) . as we will show below superpositions of fundamental boundary conditions
may also arise for rg flows in the presence of spontaneous symmetry breaking . in this paper
we investigate the rg boundaries and interfaces for the simplest minimal model - the critical ising model .
it has three primary fields : @xmath45 , @xmath46 , @xmath47 and three fundamental conformal boundary conditions with boundary states @xcite [ conf_bcs ] |= , |f= |0- | .
here @xmath48 , @xmath49 , @xmath50 stand for the ishibashi states built on the respective primaries : @xmath45 , @xmath46 , and @xmath47 .
the boundary states @xmath51 describe the fixed spin while @xmath52 describes the free spin boundary condition .
the theory has two relevant perturbations - temperature and magnetic field that are described by mass @xmath53 and magnetic field coupling @xmath54 in the free fermion theory : [ ift ] s_ift=(|+ ||+ im| ) d^2 x + h d^2 x . for real values of the couplings @xmath53
, @xmath54 all flows are massive and our results regarding the corresponding rg boundaries are summarised on the diagram presented below .
we parameterise an outgoing rg trajectory for ising field theory by points on a circle on the @xmath55-plane . for all points on the upper semicircle except for the points on the @xmath56 axis
the rg boundary condition is @xmath57 and similarly for all points on the lower semicircle we have @xmath58 .
the exceptional points are those on the @xmath56 axis .
for @xmath59 we obtain a superposition @xmath60 while for @xmath61 we obtain @xmath62 .
[ diagram_real ] ( -3,0 ) (3.3,0 ) ; ( 0,-3)(0,3 ) ; ( 0,0 ) circle [ radius=2 ] ; ( 3.6,0 ) node @xmath53 ; ( 0,3.3 ) node @xmath54 ; ( 1.8,1.8 ) node @xmath63 ; ( -1.7,1.8 ) node @xmath63 ; ( -1.7,-1.8 ) node @xmath58 ; ( 1.8,-1.8 ) node @xmath58 ; ( 3.8,1 ) ( 2.1,0.1 ) ; ( -2,0 ) node @xmath64 ; ( 2,0 ) node @xmath64 ; ( 3.8,1.2 ) node @xmath60 ; ( -2.4,0.3 ) node @xmath62 ; the fact that we obtain a superposition of boundary states for @xmath59 reflects the spontaneous symmetry breaking of the low temperature phase .
we have also investigated the imaginary magnetic field .
it is convenient to label the rg trajectories by the dimensionless ratio [ y_def ] y= .
as shown in @xcite for imaginary magnetic field when @xmath65 ( the value is taken from @xcite ) the vacuum energy is complex with a two - dimensional vacuum subspace spanned by the eigenstates with complex conjugated energy eigenvalues .
we find that in all such cases the rg boundary is @xmath60 .
for @xmath66 the vacuum energy is real and the flows are massive . we argue that in this case the rg boundary does not approach a single conformal boundary condition as we go to the far infrared . for large negative values of @xmath4
the rg flow of the vacuum vector can be well approximated by the boundary magnetic field model that is exactly solvable and provides further insight into the non - convergent behaviour . at @xmath67
the rg trajectory approaches the yang - lee edge singularity @xcite , @xcite that is described by a non - unitary conformal minimal model @xmath68 @xcite . in this case
our findings point as well to the picture in which the rg interface does not approach a single conformal interface but demonstrates an oscillatory non - convergent behaviour .
the case @xmath69 is the exceptional case of no magnetic field and the rg boundary is given by the free boundary condition as before .
we summarise these answers on the diagram below .
( -3,0 ) (3.3,0 ) ; ( 0,-3)(0,3 ) ; ( -1.41,-1.41 ) arc ( 225:135:2 ) ; ( -1.41,-1.41 ) arc ( -135:135:2 ) ; ( 3.6,0 ) node @xmath53 ; ( 0,3.3 ) node @xmath70 ; ( 1.8,2.1 ) node @xmath60 ; ( 1.8,-2.1 ) node @xmath71 ; ( -3.6,-1 ) ( -2.1,-0.1 ) ; ( -2,0 ) node @xmath64 ; ( -1.3,2.2 ) node ; ( -3.95,-1 ) node @xmath62 ; ( 0,0 ) parabola ( -1.7,2.04 ) ; ( 0,0 ) parabola ( -1.7,-2.04 ) ; the main body of the paper is organised as follows . in section 2
we work out in detail the analytically solvable case of vanishing magnetic field . in section 3
we explain the basic tcsa setup for the perturbed ising model using free fermions . in section 4
we present numerical results obtained using tcsa for real valued couplings @xmath53 and @xmath54 .
in particular we give plots for some component ratios of the type given by ( [ gen_rat ] ) .
the numerical results together with the analytic ones from section 2 lead to the diagram on fig .
[ diagram_real ] which is further discussed in section 4.2 . while tcsa works really well for the real couplings , for imaginary magnetic field when the vacuum energy is real another numerical method - truncated free fermion space approach ( tffsa ) of @xcite works much better .
this method which we explain in section 5 allows one to treat the mass coupling non - perturbatively .
in section 6 we present our numerical results obtained using tffsa for the case of imaginary magnetic field and the region @xmath72 where the vacuum energy is complex . in section 7
we discuss the region @xmath73 .
we first consider the massive flows with @xmath66 in section 7.1 . before discussing in section 7.3 the numerics for component ratios related to the flow to the yang - lee fixed point at @xmath67
we describe all conformal interfaces between the ising and yang - lee models in section 7.2 .
we finish off with a brief discussion of some open problems in section 8 .
the ising model at zero magnetic field is a theory of free fermions with euclidean action ( |+ ||+ im| ) d^2 x here the coupling @xmath53 measures the deviation from the critical temperature of the classical model : @xmath74 .
we are going to put this theory on an infinite cylinder of radius @xmath13 .
let @xmath75 be the periodic coordinate and @xmath4 - the coordinate along the cylinder .
two sectors arise - with periodic ( neveu - schwarz or ns - sector for brevity ) and anti periodic ( ramond or r - sector ) .
quantizing on the spacial @xmath76-circle we get the single - particle frequencies _
n = where @xmath77 in the r - sector and @xmath78 in the ns - sector . below we will use the index @xmath79 ( or @xmath80 ) for integers and @xmath81 ( or @xmath82 ) for half - integers . using the complex coordinate @xmath83 the mode expansion in the ns sector
is introduced as ( x , y)=_k1/2 + z ( -)^1/2 |(x , y)=_k1/2 + z ( ) ^1/2 where ( k ) = \ { r@l 1 , & k<0 + -isign(m ) , & k>0 .
we also have similar expressions in the r - sector in terms of creation and annihilation operators @xmath84 , @xmath85 which we omit for brevity .
the physical state space on a circle @xmath86 is spanned by the states [ mbasis1 ] |k_1 , , k_n_ns = b^_k_1 b^_k_n|0_ns , k_i1/2 + z with @xmath87 even and by the states [ mbasis2 ] |n_1 , , n_m_r = b^_n_1 b^_n_m|0_r , n_i where when @xmath88 the number of oscillators @xmath89 is even and when @xmath90 @xmath89 is odd .
the hamiltonian blocks are [ hff1 ] h^ns = _ n=-^ _ n+1/2b^_n+1/2b_n+1/2 + e_0^(ns ) , [ hff2 ] h^r = _ n=-^ _ nb^_nb_n + e_0^(r ) where the vacuum energies are e_0^(ns)= ( ) -|m|_-^ ( ) ( 1 + e^-|m|r ( ) ) , e_0^(r)= ( ) -|m|_-^ ( ) ( 1 - e^-|m|r ( ) ) . here
@xmath91 is a subtraction scale for the logarithmic divergence that arises from the ope of two energy operators and is responsible for the onsager singularity of free energy .
the theory with an arbitrary value of @xmath53 can be described in the @xmath92 state space using a bogolyubov transformation . the massless fermionic fields which we denote @xmath93 , @xmath94 have mode expansions & & ( z ) = ( -i ) ^1/2 _ k1/2+z a_ke^i(x+iy)k , + & & |(|z ) = ( i ) ^1/2 _ k1/2+z |a_ke^-i(x - iy)k .
the operators @xmath95 are the annihilation operators and @xmath96 , @xmath97 , @xmath98 are the corresponding creation operators .
similarly in the ramond sector the expansions are & & ( z ) = ( -i ) ^1/2 _
n a_ne^i(x+iy)n , + & & |(z ) = ( i ) ^1/2 _ n |a_ne^-i(x - iy)n where the zero modes satisfy a_0 ^ 2 = , |a_0 ^ 2 = .
we choose conventions in which the zero modes act on the spin and disorder primary states according to a_0|= | , |a_0|= -| .
the physical state space @xmath11 is spanned by the states |a_k_1^ |a_k_p^a_k_p+1^ a_k_n^|0 , k_i1/2 + z , n - in the ns sector and by the states |a_n_1^ |a^_n_qa^_n_q+1 a_n_m^| , n_i , , m - in the r sector . above @xmath99 stands for the conformal vacuum state .
the bogolyubov transformation relating the massive and massless theories is [ bogns1 ] b_k^ = f(k)\ { l@l ( |a^_k - i a_k ) , k>0 + ( a_-k^ + i |a_-k ) , k<0 . in the ns sector and [ bogr1 ]
b_n^ = f(n)\ { l@l ( |a^_n - i a_n ) , n>0 + ( a_-n^ + i |a_-n ) , n<0 .
[ bogr2 ] b_0^ = \ { l@l ( a_0 + i|a_0 ) , m>0 + ( a_0 - i|a_0 ) , m<0 .
, b_0 = \ { l@l ( a_0 - i|a_0 ) , m>0 + ( a_0 + i|a_0 ) , m<0 . in the r - sector . in the above _
k = _ k - , f(k ) = where @xmath81 is a half - integer and similar formulas hold with @xmath81 replaced by an integer @xmath79 for the r - sector .
bogolyubov transformation ( [ bogr1 ] ) , ( [ bogr2 ] ) is a proper bogolyubov transformation in the terminology of @xcite which means that it is implemented by a unitary operator @xmath100 so that u |a^_k u^ * & = & f(k)(|a^_k - i a_k ) , +
u a^_ku^*&=&f(k)(a_k^ + i |a_k ) and similarly in the r - sector with the zero modes transforming as u a_0 u^ * = a_0 , u |a_0 u^ * = sign(m)|a_0 .
the bogolyubov transformation at hand allows one to embed the states from @xmath86 into @xmath11 and vice versa .
more precisely the first embedding is realised by means of an interface operator d_m : h_0 _ m , d_m = o u where |a^_k_1
b^_-k_n |0_ns and similar relations are satisfied in the r - sector with the zero modes transforming as |a^_0 a^_0
|= |0_r .
the inverse operator d_m^-1 : h_m _ 0 , d_m^-1 = o^-1 v where @xmath101 is a unitary operator that realises the inverse bogolyubov transformation . a general formula for operators @xmath102 , @xmath103 for a given bogolyubov transformation is known ( see e.g. formula ( 5.15 ) from berezin s book @xcite ) . using it we find the following explicit formulas for the two blocks of @xmath104 : d_m^-1 = ( cc d^ns & 0 + 0 & d^r ) , [ ions ] d^ns = n()|00|_ns _ k>0 ( i(a^_k|a^_k - b_-kb_k ) + f^-1(k)(a^_kb_-k + |a^_kb_k ) ) , [ ior ] d^r = ( ) _ n=1^ ( i(a^_n|a^_n - b_-nb_n ) + f^-1(n)(a^_nb_-n + |a^_nb_n ) ) where = \ { l@l |0|_r + |0|_rb_0 , m>0 + . and all creation operators @xmath105
are understood to act on @xmath99 and @xmath106 from the left .
the normalisation factors @xmath107 , @xmath108 are given in terms of infinite products ( ) = _
n=0^ f(n+1/2 ) , ( ) = _ n=1^ f(n ) . where we denoted @xmath109 - the dimensionless mass .
taking the logarithms of @xmath107 , @xmath108 and applying abel - plana summation formula we obtain the following integral expressions [ n1 ] n ( ) = , [ n2 ] ( ) = 2 ^ 1/4 , the ratio of the normalisation factors @xmath110 interpolates between @xmath111 at @xmath112 and @xmath113 at @xmath114 . below is a plot of [ fnu ] f ( ) = this function monotonically increases from the value @xmath115 to 1 .
it passes the value @xmath116 around @xmath117 and the value @xmath118 around @xmath119 .
the interface operator can be alternatively described in terms of a local gluing condition at @xmath5 : ( x,0)=(x,0 ) , |(x,0)=|(x,0 ) .
this description is manifestly local in the @xmath76-direction . in the decompactification limit @xmath120 all excited states in the spectrum
acquire infinite energy above the vacuum .
as conformal symmetry is restored we expect the vacuum states of the infinitely massive theory to be described as conformal boundary states in the massless theory @xmath11 . for @xmath88
the system is in the low temperature ordered phase .
as @xmath53 goes to plus infinity the vacuum becomes doubly degenerate .
this happens exponentially fast in @xmath121 .
more precisely the difference of vacuum energies in the ns- and r- sectors is e ( e_0(r ) - e_0(ns ) ) = _ -^ ( ) that for @xmath122 behaves as e ~^-3/2e^- . at @xmath123
only a two - dimensional vacuum space is left in the spectrum . using ( [ ions ] ) , ( [ ior ] ) we can find the expressions for the asymptotic vacuum states . as they have infinite norm we should think of them , as we usually do for boundary states , as vectors in the dual space @xmath124 .
the details are as follows .
the normalisation factors @xmath125 , @xmath126 each goes to zero due to the casimir energy factors [ casimir ] in ( [ n1 ] ) , ( [ n2 ] ) .
this reflects the fact that the conformal boundary states to which the massive vacuum approaches have infinite norm . stripping off this vanishing factor we obtain the following asymptotic representations of the vacuum sectors of @xmath127 in @xmath11 |0_ns^+_m+ d_ns|0_ns = _ n=0^(ia^_n+1/2|a^_n+1/2)|0 , |0_r^+_m+ d_r|0_r = 2 ^ 1/4 _ n=1^(ia^_n|a^_n)| .
these states can be decomposed into conformal ishibashi states . ]
@xcite using the identification of the energy primary state [ eps_state ] |= ia^_1/2|a_1/2^|0 .
we obtain |0_ns^+ = |0 + relative normalisations to be combined into conformal boundary states . by multiplying them by @xmath128 and choosing the appropriate relative phases between the ns- and r- sector components we combine them into two cardy states which represent local boundary conditions corresponding to fixed boundary spin : @xmath57 , @xmath58 given in ( [ conf_bcs ] ) .
hence the rg boundary state that corresponds to the @xmath129 flow is a superposition @xmath130 for @xmath90 the system is in the high temperature disordered phase . in the r - sector
the lowest energy state is the fock space one - particle excitation @xmath131 so that all r - sector states disappear from the spectrum as @xmath132 .
the remaining vacuum state is |0_ns^-_m- d_ns|0_ns = _ n=0^(-ia^_n+1/2|a^_n+1/2)|0 , which as above can be decomposed into the ishibashi states as |0_ns^- = |0- | that already corresponds to a normalised cardy boundary state @xmath133 giving the free boundary condition .
thus in this case |rg= |f . in the presence of magnetic field ( the case we take up in the next section )
the main quantities we will calculate numerically are component ratios of the type ( [ gen_rat ] ) discussed in the introduction . specifically for the massive flows of the ising field theory
it is convenient to introduce two ratios : [ ising_ratios ] _
here @xmath44 stands for the vacuum of perturbed theory as it appears inside the critical ising state space .
it is instructive to look at these component ratios for the @xmath56 flows where we can get the exact formulas for them .
the ( unnormalised ) vacuum of the perturbed theory ( at finite value of @xmath53 ) is @xmath134 that always lies in the ns - sector . from ( [ ions ] )
^0 = = = sign(m ) g ( ) where [ gnu ] g()= - .
the function @xmath135 asymptoticaly goes to 1 very slowly , as a power function .
here is a plot of @xmath135 [ g_fig ] for zero magnetic field , @xmath88,title="fig : " ] this function monotonically increases and becomes larger than @xmath116 past @xmath136 and larger than @xmath118 past @xmath137 .
these values of the dimensionless mass are very large .
this shows that it may be much harder to read off the component ratios from tcsa numerics than say the energy spectrum for which the convergence is typically exponential .
the further we have to go into the infrared the greater the tcsa errors can be for a fixed truncation level .
we will see that while the perturbative corrections grow the numerics demonstrates that the total truncation error remains bounded .
for @xmath129 the first excited vector remains in the spectrum .
it is @xmath138 for which we can define a ratio of components
| | = g ( ) that again converges as slowly as @xmath139 .
we would like now to consider a generic perturbation of the critical ising model with @xmath140 and @xmath141 . for real - valued @xmath141
the system flows to a trivial theory with a single vacuum .
we can study numerically the ratios @xmath142 using tcsa .
we work in the massless fermion physical space @xmath11 .
the perturbed euclidean action of the ising field theory is given in ( [ ift ] ) .
let us choose an orthonormal basis in @xmath11 described in section [ preliminaries ] . in the ns sector
we choose [ basis1 ] ||k_1 , |k_p ; k_1 , k_q_ns = > 0 , p+q - and |k_1 > |k_2 >
> |k_p , k_1 > k_2 >
> k_q . in the r - sector
we choose [ basis2 ] ||n_1 , |n_p ; n_1 , n_q_r = n_1 , n_q_r = |a_|n_1^ |a^_|n_pa^_n_1 a_n_q^| , p + q - . here
we assume n_i , |n_j , n_i>0 , |n_j>0 and |n_1 > |n_2 >
> |n_p , n_1 > n_2 >
> n_q . we are going to work in a finite - dimensional truncated space @xmath143 is spanned by ( [ basis1 ] ) , ( [ basis2 ] ) , ( [ basis3 ] ) satisfying the following additional constraints _
i=1^p|k_i = _
i=1^qk_i n_c , _
i=1^p|n_i = _
i=1^qn_i n_c where @xmath144 is the integer that sets the level truncation truncates the total virasoro weight . note that in some tcsa schemes the truncation is done in the level of descendants only rather than in the total weight . ] .
we choose the space @xmath143 to contain only zero spin vectors that is relevant for identifying rg boundary states . [ cols="^,<,<,<,<",options="header " , ] in operator quantisation the matrix elements of @xmath47 factorise into holomorphic and antiholomorphic factors [ sigma_matrix ] & & _ ns|k_1 ,
|k_p , k_1 , k_q|(0,0)||n_1,
|n_r , n_1 , , n_s _ r + & & = ( ) ^1/8(-1)^q(q-1)/2 + s(s-1)/2 + r(s+1 ) g(|k_1 ,
|n_1 , , |n_r ) + & & g(k_1 , , k_q , n_1 , , n_s ) where [ g ] & & g ( k_1 , k_p , n_1 , ,
j=1^p g_ns ( k_j ) _ i=1^rg_r ( n_i ) ( _ 1i < jp ) + & & ( _ 1i
< j q ) ( _ 1i p ; 1jr ) where g_ns(k ) = , g_r(n ) = are the massless leg factors .
the antiholomorphic factor @xmath145 is given by the same formula as ( [ g ] ) with @xmath146 mode numbers replaced by @xmath147 .
the rg trajectories of the ising field theory ( [ ift ] ) are labeled by a dimensionless parameter y= for which we use the same notation as @xcite .
we also use a dimensionless distance scale r = r|h|^8/15 . in terms of these two scaling variables
the dimensionless mass can be expressed as = r|m| = |y| r .
the tcsa hamiltonian is [ h_tcsa ] h_tcsa = .
here @xmath9 , @xmath148 and @xmath89 are matrices not containing any dimensional parameters or scaling variables @xmath4 , @xmath149 .
they are defined as follows .
we have h_0 = _ n=0^ ( n + 1/2)[a^_n+1/2a_n+1/2 + |a^_n+1/2|a_n+1/2 ] - in the ns - sector and h_0 = _ n=0^ n [ a^_na_n + |a^_n|a_n ] + in the r - sector .
the matrix @xmath148 has the same matrix elements as ( [ sigma_matrix ] ) with the factor @xmath150 removed .
the matrix @xmath89 is the matrix for the operator @xmath151 in the basis given by ( [ basis1 ] ) , ( [ basis2 ] ) , ( [ basis3 ] ) . if @xmath152 is an eigenvalue of @xmath153 we also define dimensionless eigenvalues e(r ) = e(r ) and e(r ) = . for a flow to yang - lee fixed point the differences of the eigenvalues @xmath154 interpolate between the differences of scaling dimensions at the fixed points
( see a detailed discussion in section [ ycr_sec ] ) . to identify rg boundaries for the massive flows we are going to calculate numerically the component ratios @xmath155 , @xmath156 defined in ( [ ising_ratios ] ) .
also later we will consider the rg flow to yang - lee fixed point which occurs at imaginary magnetic field . in that case
we will also look at the ratios [ ising_ratios2 ] _
^i = , _ ^i = corresponding to the @xmath157-th excited energy eigenstates @xmath15 , @xmath158 .
tcsa results at finite @xmath159 of course differ from the exact renormalised qft results at @xmath160 .
some of the corrections correspond to redefining the coupling constants and can be taken into account using a certain rg scheme @xcite , while there are also non - local corrections @xcite , @xcite .
the redefined coupling constants corresponding to the rg scheme of @xcite can be approximated by running couplings that depend on continuous @xmath161 .
let @xmath162 .
the rg equations for the running couplings @xmath163 , @xmath164 then read = -ar , = -sbt^7/4r^15/4 where a = 0.01 , b=0.004 and @xmath165 for real magnetic field and @xmath166 for the imaginary one . these equations
are to be solved with the bare couplings from ( [ h_tcsa ] ) taken as initial condition at @xmath167 .
both couplings change quite slowly .
for example for @xmath168 and @xmath166 ( close to the yang - lee trajectory ) we find numerically for @xmath169 that for @xmath170 the effective values are @xmath171 , @xmath172 .
the mass coupling @xmath173 changes particularly slowly .
the method for dealing with non - local perturbative corrections to energy eigenvalues has been worked out in @xcite , @xcite .
one strategy to implement such corrections in practice is by varying @xmath159 and by fitting the variations in numerical results to combinations of negative powers of @xmath159 . for the ising field theory
this has been done in @xcite , @xcite ( in particular see appendix b in @xcite where rg corrections and fittings in both tcsa and tffsa are discussed in detail ) .
perturbative truncation corrections to eigenvectors ( which would be of interest for calculating the component ratios @xmath174 ) have not been so far systematically investigated .
moreover as figure [ g_fig ] shows one can expect a very slow approach of @xmath174 to their ir asymptotics that does not make perturbative corrections useful for finding their ir values .
nevertheless our tcsa numerical results , that we present in the next section , show that the component ratios for real magnetic field asymptotically approach constant values with ( non - perturbative ) truncation errors staying bounded .
since the running couplings corrections are really tiny and do not change any qualitative conclusions and also to make direct comparison with the results of @xcite , @xcite , @xcite possible , everywhere in the paper we present uncorrected ( raw ) tcsa and tffsa data .
the component ratios we are going to find numerically are @xmath155 and @xmath156 defined in ( [ ising_ratios ] ) . the sign of @xmath156 is correlated with the sign of @xmath54 . for simplicity
we will always choose @xmath54 so that @xmath156 is positive . when a magnetic field perturbation is present it is natural to expect the @xmath175 component to be present in the rg boundary state .
as we do nt expect any vacuum degeneration we expect the rg boundary state to be @xmath57 or @xmath58 depending on the sign of the coupling @xmath54 .
this means that @xmath176 should approach 1 and @xmath177 should approach @xmath178 .
indeed this is what we observe in tcsa numerics presented below for a range of values of @xmath4 .
the approach to asymptotic values is always quite slow - of the type the exact solution for @xmath139 demonstrates in the @xmath56 case . on the graphs for @xmath179 below the red dashed line corresponds to the value @xmath113 .
the results presented below are not very sensitive to the the cutoff level @xmath144 ( see a more detailed discussion below ) . for illustration we chose @xmath180 .
] ] ] ] ] ] ] ] we observe that the ratio @xmath179 converges to its expected limiting value with much better accuracy than @xmath139 . while for the values of @xmath181 we have @xmath182 for @xmath139 typical asymptotic values are near @xmath116 . in the disordered phase , where @xmath27 , the mass term dominates at small distances so that @xmath139 starts out with negative values .
the magnetic field takes over at sufficiently large distances with @xmath139 passing through zero and monotonically increasing to a positive asymptotic value that is less than 1 .
thus for @xmath27 the convergence of @xmath139 to its asymptotic value is slower than in the ordered phase .
it should be noted that at the large values of @xmath149 we considered : @xmath183 , the values of @xmath184 and @xmath185 are large .
this means that any perturbative corrections are large in that region .
however the numerics shows that the perturbative corrections must sum up to a form @xmath186 where @xmath187 is a bounded function of @xmath149 .
the following plot shows how the numerical values of @xmath139 taken for @xmath188 change when we change the truncation level @xmath159 .
numerics at @xmath188 for different values of @xmath144 ] it is not clear to us why the total truncation correction remains bounded for such large values of the scale but this is certainly crucial for getting reasonable numerical results for the ratios @xmath190 due to their extremely slow asymptotics .
we focused on two particular ratios @xmath191 for convenience and for practical reasons ( these ratios are related to the lowest weight components and we expect them to be best approximated by tcsa ) .
but in general we expect all components of the asymptotic vacuum vector to arrange themselves according to the components of cardy boundary states ( or their superposition ) . to get a better measure of that we can plot all vacuum components .
more precisely let @xmath192 , @xmath193 stand for the basis elements in the critical ising model that correspond to scaling fields of dimension @xmath194 , @xmath195 in the ns and r - sectors respectively .
up to phases the basis elements are given by formulae ( [ basis1 ] ) , ( [ basis2 ] ) , ( [ basis3 ] ) .
we fix the phases by state - operator correspondence ( see ( [ eps_state ] ) and label these basis vectors in such a way that the conformal weights are monotonically non - decreasing : @xmath196 if @xmath197 and similarly in the r - sector .
let @xmath198 be the vacuum vector components relative to this basis : [ conf_basis ] |0= _ i c^ns(i)|i_ns + _ j c^r(j)|j_r .
for @xmath199 we have 556 basis vectors in the ns sector out of which 92 are diagonal . on the graph below all components
are plotted , out of which only the diagonal ones are clearly visible while the rest of the components are much smaller and on the graph they are concentrated on the @xmath157-axis .
we see that the diagonal components have the same phase and decrease in amplitude with increasing conformal weight practically forming a continuous curve .
the decrease is a truncation effect . in the untruncated theory
we would expect all components to be the same according to the composition of the cardy states ( [ conf_bcs ] ) . , @xmath200 , @xmath180 .
the red numbers mark the conformal weights of the components . only diagonal components are distinguishable above the @xmath157-axis . ]
the ramond components form a similar pattern with the diagonal components decoupled from the much smaller non - diagonal ones .
, @xmath200 , @xmath180 .
the red numbers mark the conformal weights of the components . ]
another way to see the domination of the diagonal states in the vacuum is by calculating the share of the diagonal states in the square of the norm : s= .
we found that numerically @xmath201 is above @xmath202 for a range of @xmath4 , @xmath159 and @xmath149 between @xmath203 and @xmath204 . below is a sample graph , @xmath199 . ] as can be seen from the table given on fig .
[ table ] the number of diagonal states grows significantly slower with @xmath144 than the total number of states .
it is remarkable that these states dominate the vacuum with such a high proportion .
even more unexpectedly we found that the first and second excited states have about the same high proportion of the diagonal states in their norm .
it remains to be seen whether this observation can be put to use to improve the numerics .
our analytical results for the vanishing magnetic field together with the numerical results for measuring @xmath205 are summarised on the diagram depicted on fig .
[ diagram_real ] .
we observe that the two exceptional rg boundaries : @xmath60 , @xmath62 , are unstable .
there are boundary rg flows from each of them that end up either with @xmath57 or with @xmath58 .
these end points are precisely the rg boundary conditions corresponding to the one - dimensional regions joined at the exceptional points .
this fact has the following physical explanation .
suppose we are at an exceptional point that belongs to a submanifold separating higher - dimensional regions labeled by different rg boundaries .
far in the infrared the rg interface is almost non - transparent and can be well approximated by a conformal boundary condition .
if we add now a small bulk perturbation on the massive side of the interface that moves us away from the separating submanifold this will result in an effective perturbation of the rg boundary and a subsequent effective rg flow to a new rg boundary condition .
( more precisely this is still a bulk plus boundary flow , however in this representation low energy degrees of freedom survive only on the boundary so it is effectively described by a pure boundary flow . ) in the ising field theory case we can be more quantitative in describing these effective boundary rg flows that start from the exceptional rg boundaries .
these flows are triggered by the @xmath47-perturbation taken at a large dimensionless mass @xmath121 .
the matrix elements of @xmath47 in a massive fock space are given explicitly in formula ( [ matrix_mass ] ) .
in the limit @xmath206 all of these matrix elements go to zero or to a finite constant except for the vacuum - vacuum and the vacuum - one - particle ones that diverge .
this means that for @xmath88 effectively the @xmath47 perturbation at large mass will act as a boundary identity field that mixes the two vacua , while for @xmath90 the one - particle matrix elements give rise to a boundary magnetic field perturbation .
in this section we outline how the truncated free fermion space approach ( tffsa ) invented in @xcite can be used to find numeric approximations to the ratios @xmath207 .
this method has the advantage over the tcsa in treating the mass coupling non - perturbatively and also in having greater control over the large @xmath149 asymptotics .
however it has its own subtleties related to uv divergences which we are going to discuss as well . in tffsa
one uses the massive fermion physical space @xmath86 described in section [ preliminaries ] .
we write the ising field theory hamiltonian as [ ham2 ] h = h_ff + h_0^r ( x,0 ) dx where @xmath208 is the free massive fermion hamiltonian given in ( [ hff1 ] ) , ( [ hff2 ] ) and the matrix elements of the magnetic field perturbation are given by @xcite [ matrix_mass ] & & _ nsk_1 , k_2 , , k_n|(0,0)|n_1,n_2 , ,
n_m_r= i^ |s(r ) _
i=1^ng(_k_i ) _ j=1^m
g(_n_j ) + & & _ 1i
< j < n ( ) _ 1p < q < m
( ) + & & _ 1rn ; 1sm ( ) . here @xmath209 are finite size rapidities in the r- and ns - sectors : ( _ n ) = , n ( _ k ) = , k + z , the functions @xmath210 are the leg factors defined as g ( ) = , g ( ) = where ( ) = _ -^+ .
the overall factor @xmath211 in ( [ matrix_mass ] ) is the vacuum - vacuum matrix element |s(r)= \
{ l@l _ ns0|(0,0)|0_r , & m>0 , + _
ns0|(0,0)|0_r , & m<0 . .
we have the following explicit expressions |=
|m|^1/82 ^ 1/12e^-1/8a^3/2 where @xmath212 is glaisher - kinkelin constant , and s(r ) = ( ) ^2 d_1d_2 .
the hamiltonian ( [ ham2 ] ) is restricted to a truncated space @xmath213 that is spanned by vectors ( [ mbasis1 ] ) , ( [ mbasis2 ] ) satisfying _
i=1^n k_i = _ j=1^mn_j = 0 , and _
i=1^n|k_i| 2n_c , _
j=1^m |n_j| 2n_c .
here @xmath144 is an integer that controls the truncation .
in contrast with the tcsa , it no longer is related to the energy of the unperturbed hamiltonian @xmath208 .
the total dimensions of truncated spaces are the same as the ones given in figure [ table ] .
while the tffsa eigenvectors lie in @xmath86 we can use the interface operator @xmath214 to obtain their image in @xmath11 .
this can be formally thought of as combining ( or fusing ) two perturbation interfaces : the mass interface @xmath215 and the magnetic field interface @xmath216 .
the last one corresponds to perturbing the free massive theory by the magnetic field .
( 3.4,0 )
( 5.8,0 ) arc ( 270:90:0.6 and 1.1 )
( 3.4 , 2.2 ) arc ( 90:270 : 0.6 and 1.1)cycle ; ( 5.8,0 ) ( 9,0 ) arc ( -90:90:0.6 and 1.1 ) ( 5.8 , 2.2 ) arc ( 90:270 : 0.6 and 1.1)cycle ; ( 0,0 ) arc ( 270:90:0.6 and 1.1 ) ; ( 0,0 ) arc ( -90:90:0.6 and 1.1 ) ; ( 0,0)(9,0 ) ; ( 0,2.2 cm ) (9 , 2.2 cm ) ; ( 9,0 ) arc ( 270:90:0.6 and 1.1 ) ; ( 9,0 ) arc ( -90:90:0.6 and 1.1 ) ; ( 3.4,0 ) arc ( 270:90:0.6 and 1.1 ) ; ( 3.4,0 ) arc ( -90:90:0.6 and 1.1 ) ; ( 5.8,0 ) arc ( 270:90:0.6 and 1.1 ) ; ( 5.8,0 ) arc ( -90:90:0.6 and 1.1 ) ; ( 3.6,-0.35 ) node @xmath217 ; ( 6.0,-0.4 ) node @xmath218 ; ( 1.5,1.2 ) node @xmath219 ; ( 1.5 , 0.85 ) node @xmath56 ; ( 4.6,1.2 ) node @xmath140;(4.65,0.85 ) node @xmath220;(7.5,1.2 cm ) node @xmath221 ; ( 7.5,0.85 cm ) node @xmath222 ; ( 3.4 , 2.25 ) ( 3.4,2.6 ) ; ( 5.8 , 2.25 ) ( 5.8,2.6 ) ; ( 3.4 , 2.45 ) ( 5.8,2.45 ) ; ( 4.55,2.64 ) node @xmath46 ; the mass interface operator was constructed analytically in section [ mass_interface ] while we can use the tffsa numerics to obtain information on the second interface .
fusion of conformal interfaces usually contains multiplicative divergences ( see e.g. the discussion in @xcite , @xcite ) .
this divergence is regulated by the truncation of @xmath86 present in tffsa .
however one may still worry whether the fusion procedure gives the same interface as the tcsa one in the @xmath223 limit .
while we can not exclude this situation with definiteness we have not observed anything in the numerical results to be presented below that would suggest this scenario .
let @xmath224 be the projector implementing tffsa truncation at level @xmath144 .
then we can write for the component ratios ( [ ising_ratios ] ) , ( [ ising_ratios2 ] ) _
^i = . here and below @xmath225 . using ( [ ions ] ) , ( [ ior ] ) we calculate _
^i= -sign(m ) , _ ^i = 2 ^ 1/4f ( ) where @xmath226 and @xmath135 are given by ( [ fnu ] ) , ( [ gnu ] ) , and t_1(v,)= _ ns0| ( -i g ( ) b_-1/2b_1/2 ) _ n=1^n_c-1 ( 1 - i b_-n-1/2b_n+1/2 ) |v , t_2(v,)= _ ns0| _ n=1^n_c-1 ( 1 - i b_-n-1/2b_n+1/2 ) |v , u(v , ) = _ r0| ( b_0)^p _ n=1^n_c ( 1 - i b_-nb_n ) |vwhere @xmath227 if @xmath27 and @xmath228 when @xmath229 .
we see from these expressions that part of the scale dependence of @xmath230 comes from the functions @xmath226 and @xmath135 and from @xmath231 which we know analytically . in practice
we observed a faster rate of convergence of @xmath232 to its asymptotic value than that of @xmath233 .
this can be at least partially attributed to the faster convergence of @xmath226 than that of @xmath135 to its asymptotic value 1 .
having noted this we can formally define quantities in which @xmath234 , @xmath235 are taken at @xmath236 : _ ^i =- sign(m ) , _
^i = 2 ^ 1/4 . in practice
we found that using @xmath237 gives a small improvement in the convergence rate towards the asymptotic values .
as these quantities do not have a clear physical meaning ( the eigenvector @xmath15 is still taken at finite @xmath121 ) we are not going to present the numerical results for them in the paper . applied to the flows triggered by real @xmath54 the tffsa method here described gives data very similar to the tcsa one presented in section [ tcsa_section ] .
the method however has a significant advantage when applied to the imaginary magnetic field flows discussed in the forthcoming sections .
the ising field theory ( [ ift ] ) taken at imaginary values of @xmath54 is not unitary .
however the corresponding hamiltonian @xmath12 enjoys the following symmetry [ ssym ] shs = h^ where @xmath201 is the operator that multiplies any ramond sector vector by @xmath238 and leaves any ns sector vector intact . to see the implications of this symmetry
consider @xmath12 as an operator in @xmath11 and choose a basis in which @xmath12 is ( complex ) symmetric .
for example we can take the conformal basis described before equation ( [ conf_basis ] ) . in this basis the matrices @xmath89 and @xmath148 in ( [ h_tcsa ] ) are symmetric and hence @xmath12 is symmetric as well . as before denote these basis vectors as @xmath239 , @xmath240 .
if @xmath12 with eigenvalue @xmath241 then ||v__i with eigenvalue @xmath242 .
( the bar stands for complex conjugation everywhere . )
then ( [ ssym ] ) in this particular basis implies that hs||v_= |s||v_that is |v_|_i |c^ns(i)|i_ns - _ j |c^r(j)|j_r is an eigenvector of @xmath12 with eigenvalue @xmath242 .
thus the energy eigenvalues are either real or form a pair of complex conjugated values .
moreover the above implies that if the vacuum of @xmath12 has real energy and is non - degenerate then the vacuum vector must be of the form |v_0= |0 + c^ns|+ ic^r|+ where @xmath243 , @xmath244 are real and we show only the three lowest components .
hence in this case @xmath139 is real and @xmath245 is imaginary .
alternatively if the vacuum energy is complex and the vacuum space is a two - dimensional subspace corresponding to two conjugate eigenvalues then the corresponding eigenvectors can be written as [ pair_eig ] & & |v_0= |0 + c^nse^i|+ c^r e^i |+ + & & ||v_0= |0 + c^nse^-i|- c^r e^-i |+ where @xmath243 , @xmath244 are real and hence @xmath246 @xmath247 are both complex in general
. since none of the conformal boundary states ( [ conf_bcs ] ) has a complex value of @xmath248 we see for conformal boundary states can be seen as a consequence of locality of the boundary condition . ]
that in the first case _ as long as the vacuum eigenvalue remains real the vacuum vector has no chance of approaching a conformal boundary state as we move along an rg trajectory .
_ in the second case the only way @xmath249 and @xmath248 can become real at the end of the rg flow is if the phases @xmath250 , @xmath251 tend to 1 or -1 as @xmath252 .
if this is the case the eigenvectors given in ( [ pair_eig ] ) should either tend to the pair of fixed boundary states @xmath51 or to @xmath253 .
we will present evidence that supports the first possibility .
we now focus on the case of complex vacuum energy .
the spectrum of the ising field theory for imaginary magnetic field and arbitrary mass was investigated in @xcite , @xcite , @xcite using tffsa as well as analytic results .
it was shown in those papers that for @xmath72 where @xmath254 the vacuum energy becomes complex for sufficiently large values of @xmath149 .
the vacuum vectors form a pair with conjugated energy values .
as discussed in @xcite for positive @xmath4 and large enough @xmath149 the vacuum and excited states form complex pairs with @xmath255 asymptotically approaching constant values and @xmath256 asymptoting to straight lines with equal slopes ( free energy density ) thus furnishing particle - like excitations with complex masses . here
we present plots of the numerical results ( using tffsa ) for the ratios @xmath249 , @xmath248 calculated for the two vacuum vectors at @xmath257 . for the two vacuum vectors at @xmath257 , @xmath258 . ] for the two vacuum vectors at @xmath257 , @xmath258 . ] for the two vacuum vectors at @xmath257 , @xmath258 . ] for the two vacuum vectors at @xmath257 , @xmath258 . ] comparing these plots to the ratios in conformal boundary conditions ( [ conf_bcs ] ) we find that the rg boundary is a superposition @xmath60 .
the case of large positive @xmath4 can be understood in terms of boundary rg flows ( see the discussion in section [ massive_summary ] ) .
this regime corresponds to starting with a large @xmath88 and adding to it a small imaginary magnetic field perturbation .
thus with a good approximation we start with a vacuum described by the conformal boundary condition @xmath60 and the perturbation is just the identity field between the two boundary conditions multiplied by an imaginary coupling .
the only effect of this is a relative phase factor which keeps rotating as we change the scale but does not change any physical quantities .
we have also checked numerically the region @xmath259 .
as long as the vacuum complex pair does not experience collisions with upper level eigenvalues the picture is qualitatively the same as presented on the above plots .
such collisions are sensitive to the truncation level so strictly speaking we do not have a proof that the above picture remains in the extreme @xmath252 asymptotics at infinite truncation level .
but it looks highly plausible to us . to summarise the numerical results support the rg boundary being @xmath60 for all cases when the vacuum energy is complex and we have a pair of vacuum vectors .
for @xmath73 numerical studies @xcite show that the vacuum energy remains real until a large positive value if @xmath149 . the latter value at which the vacuum eigenvalue collides with the first excited eigenvalue and forms a conjugate pair is numerically the larger the larger @xmath144 is and presumably such collision is a numerical artefact .
we found that tffsa method works much better than tcsa in this regime .
in particular the lowest eigenvalue stays real for much larger value of @xmath149 when using the tffsa method .
we start discussing the numerical results for this region by taking a large negative @xmath4 where the vacuum energy stays real for large @xmath149 s and where we can use the boundary rg picture discussed in section [ massive_summary ] .
below we present numerical data for a sample point @xmath260 .
one finds that the behaviour of the component ratios is very different here from the situations discussed before .
namely the overlap of the perturbed vacuum with the unperturbed one goes to zero and both @xmath155 and @xmath156 go to infinity at a finite value of rg scale @xmath149 .
the ratio @xmath261 remains finite .
define t= , = || . here is a plot of these ratios for @xmath260 [ data1 ] and @xmath179 at @xmath260 , @xmath199 . ] and @xmath179 at @xmath260 , @xmath199 . ]
[ data2 ] and @xmath262 at @xmath260 , @xmath199 . ] and @xmath262 at @xmath260 , @xmath199 . ]
we find that @xmath263 vanishes and @xmath264 blow up at @xmath265 .
the ratio @xmath262 remains finite at this point with the value @xmath266 .
for large negative values of @xmath4 we can approximate the flow of the vacuum vector by an rg boundary flow .
for large negative values of @xmath53 the vacuum is well approximated by the boundary state @xmath62 corresponding to free boundary spin . switching on a small imaginary magnetic field in the bulk
perturbed this boundary condition by an imaginary boundary magnetic field .
the critical ising model with a boundary magnetic field is a gaussian theory and can be solved exactly @xcite , @xcite .
it is not hard to extend this solution to imaginary magnetic field .
the action functional on a cylinder for this model reads ( |+ || ) d^2 x + ( |+ aa + h_b(+ || ) a ) dy where the boundary is located at @xmath267 , @xmath268 and the boundary magnetic field coupling is taken here to be @xmath269 with @xmath270 - real .
the boundary fermion @xmath271 accounts for the double degeneracy of the vacuum . adopting the boundary state found in @xcite for the case of imaginary magnetic field
we obtain [ b_state_chat ] & & |h_b= e^-(r)where @xmath272 and the sign in front of the ramond component is fixed by the sign of @xmath270 .
this exact solution gives us the following component ratios [ gexactc ] _
^0= i 2 ^ 1/4 , [ gh ] = .
the vacuum component of @xmath273 vanishes linearly at @xmath274 that is @xmath275 .
this results in @xmath155 and @xmath156 having a simple pole at this point while their ratio is finite and given ( in absolute value ) by function ( [ gh ] ) a portion of which which we depict below for imaginary boundary magnetic field model . ] comparing ( [ gexactc ] ) , ( [ gh ] ) with fig [ data1 ] and fig [ data2 ] we find qualitatively exactly the same picture .
we find that the phases of all ratios are the same as in ( [ b_state_chat ] ) and that at @xmath276 in the ns sector all components are small except for the components including the @xmath277 oscillators as in ( [ b_state_chat ] ) .
since we do not know how the effective boundary @xmath270 depends on @xmath4 and @xmath149 we can not match the two curves for @xmath262 ( they are parameterised differently ) .
however we can still match quantitatively the two special values : @xmath278 - the minimal value and @xmath279 - the value at the point at which @xmath264 blow up .
the numerical tffsa values : @xmath280 , @xmath281 are approximately 1.573 and 2.044 respectively .
these values are within few percent from those predicted by the boundary magnetic field model .
the match gets better for larger values of @xmath282 and larger @xmath144 . in the numerics for @xmath260
the data continues past the blow up point @xmath283 with @xmath262 continuing to grow . in the boundary magnetic field model going past the first special point @xmath284
the ratio @xmath262 blows up at @xmath285 . in the numerics we do not quite get to that second point as the vacuum and the first excited energy levels collide and form a complex pair .
the collision point is sensitive to the truncation level @xmath144 and is moved towards larger values of @xmath149 as @xmath144 is increased .
so presumably in the untruncated theory the vacuum remains real at all scales .
as explained at the beginning of section [ complex_vac_sec ] as long as the vacuum energy remains real and @xmath179 does not vanish ( and thus remains imaginary ) we can not approach a local conformal boundary condition .
so , what happens
? given a good match at the onset of the flow with the imaginary boundary magnetic field model we are going to rely on it in describing what happens as we continue increasing the scale .
as we can clearly see from ( [ b_state_chat ] ) the model keeps going through a sequence of special points at which @xmath286 is a ( positive ) integer or half integer . at half
integer points @xmath287 all boundary state components with level less than @xmath288 in the ns sector vanish and in the higher weight components only those including @xmath289 oscillators survive . for @xmath79 large the low energy components in the r - sector approach those of the @xmath175 ishibashi state . at integer points @xmath290 similarly the low energy components in the r - sector are wiped out while those in the ns - sector approach those of the @xmath291 ishibashi states combination .
if we focus on the low lying components with level smaller than @xmath292 then asymptoticaly ( up to an overall factor ) we have |h_b~ ( ) [ |0 + | ] i
( ) 2 ^ 1/4| + where the ellipsis stands for terms that contain components of level larger than @xmath292 .
thus as @xmath292 goes to infinity we will see a never ending rotation of the two combinations of ishibashi states .
this cyclic behaviour is of course in violation of @xmath293-theorem which however is possible here because we are in a non - unitary situation . at any real value of @xmath54
the ising field theory flows to a trivial fixed point . for pure imaginary @xmath54 with a certain value of the scaling parameter @xmath67 the theory flows to the yang - lee fixed point which is a non - unitary minimal model @xmath68 with central charge @xmath294 .
it has two primary fields : the identity and a field @xmath295 with scaling dimension @xmath296 and correspondingly two conformal boundary conditions with cardy states @xmath297 .
the value of @xmath298 was most recently estimated numerically @xcite to be @xmath299 .
all conformal interfaces between the ising and yang - lee models are known @xcite due to a remarkable fact that the tensor product @xmath300 is itself a minimal model @xmath301 with @xmath302 modular invariant .
the interfaces are described as boundary conditions in the tensor product theory @xmath301 which were found in @xcite .
the @xmath301 theory has 12 primary states and thus 12 elementary conformal boundary conditions .
we will use the same conventions as in @xcite , @xcite in which chiral primaries @xmath303 are labelled by @xmath304 and @xmath305 - the set of @xmath306 exponents .
the tensor products of primaries are identified as = _ 1,1 , = _ 1,5 , = _ 1,4 , = _ 3,7 , = _ 3 , 5 , = _ 3,8 . the conformal boundary states @xmath307 are labelled by a pair @xmath308 , @xmath309 ( the labels of nods of the @xmath306 dynkin diagram ) .
the decomposition into the ishibashi states can be written as ) was derived in @xcite with the assumption that the matrix @xmath310 is unitary . a more general derivation in which this assumption was not made
was later presented in @xcite . ] @xcite [ cardy_e6 ] |= _ r , s |(r,s ) where [ psi ] _
r , a^(r,s ) = s_rr^s_a and @xmath311 is the matrix made of the eigenvectors of the @xmath306 adjacency matrix .
explicitly we have [ matrix ] ( ^s_i ) = ( rrrrrr a & & b&b & & a + b & & a & -a & - & -b + c&0&-d&-d & 0 & c + b & - & a&-a & & -b + a&- & b & b & - & a + d & 0 & -c&c & 0 & -d ) where a = , b= , c = , d= . in ( [ matrix ] ) the row index
@xmath312 labels the boundary states and the column index @xmath305 labels the primaries .
the @xmath201-matrices present in ( [ cardy_e6 ] ) , ( [ psi ] ) are s_rs , r s = ( -1)^(r+s)(r+s ) ( r r ) ( s s ) , s_rr = ( rr ) . using the above we calculate the component ratios
@xmath233 , @xmath232 as ratios of one - point functions of factored primaries in the @xmath306 theory . as they turn out to be independent of the first index - @xmath149 labelling the boundary states , we present the answers in the form of 6-vectors with components labeled by the second index - @xmath313 . [ interfaceg1 ]
_ ^0 = = ( r 1 + -1 + 1 + -1 + 1 + -1 ) , _ ^0 = = ( r + + 0 + - + - + 0 ) , [ interfaceg2 ] _ ^1 = = ( r + 1 + -1 + 1 + + - ) , _ ^1 = = ( r + + 0 + - + - + 0 ) where = 2 ^ 1/41.189 , = 2 ^ 1/4()^1/20.615 , = 2 ^ 1/4()^1/2 2.297 , = 3.732 .
we observe that knowing the above four component ratios determines the index @xmath313 uniquely .
furthermore the factorised interfaces correspond to @xmath314 .
since a magnetic perturbation is present on the ising to yang - lee trajectory we expect a non - vanishing @xmath106-component in the vacuum .
this means that there are essentially two types of candidates for the ising - yang - lee rg interface - the factorised defects @xmath315 and the non - factorisable defects corresponding to @xmath316 .
note that the pair @xmath316 forms a doublet under @xmath317 spin reversal symmetry so that there is essentially one non - factorizable candidate interface . to determine the index @xmath149 one would need to know other ratios of one - point functions e.g. the ratio = i ( ) ^3/4 , = -i ( ) ^1/4 which is independent of @xmath313 .
however this requires knowing how the ising eigenvector @xmath318 is realised in the yang - lee state space that is inaccessible by tcsa method which only gives the infrared theory eigenvectors in the uv theory space . in the next section
we will argue that the rg interface for the ising to yang - lee flow does not asymptote to any single conformal interface . as a final remark about the conformal interfaces considered in this section we note that using the cardy constraint we can also find the spectrum of interface fields .
we found that all interfaces contain relevant fields ( and are thus unstable ) except for the factorizable ones : @xmath319 .
the approach to a non - trivial fixed point at @xmath67 is marked by the appearance of a `` nose '' on the plot of dimensionless energies @xmath320 depicted below : the vacuum and the first excited level energies get asymptoticaly close without collision .
the relative energies @xmath321 level to constant values equal to the conformal dimension difference @xmath322 .
we find numerically ( see the second graph below ) that at truncation level @xmath199 the difference @xmath323 levels at the value @xmath324 while @xmath325 approaches @xmath326 .
these values approximate the dimension gaps for the yang - lee operators @xmath45 and @xmath327 respectively ( the lowest dimension state is @xmath295 with dimension -0.4 ) .
[ nose ] , @xmath199 . ] , @xmath199 . ] for the truncation level @xmath180 we find numerically that the lowest two levels merge into a complex pair at @xmath328 .
looking at the behaviour at different values of @xmath159 we find that the merging point moves to higher @xmath149 as we increase @xmath144 .
so as in the case @xmath66 such a merger seems to be merely a finite precision artefact which however limits the domain of @xmath149 we can investigate numerically . in the previous section we found explicitly the component ratios @xmath264 , @xmath329 for all conformal interfaces between the critical ising and yang - lee models ( see formulai ( [ interfaceg1 ] ) , ( [ interfaceg2 ] ) ) .
crucially all possible values are real while , as discussed in section [ complex_vac_sec ] , for as long as an eigenvalue ( vacuum or excited ) stays real the corresponding ratio @xmath249 is real and @xmath248 is imaginary .
this leaves us with two possibilities : _ either the limiting conformal interface does not exist as in the case of massive flows discussed before , or the rg interface approaches a conformal interface with _ @xmath330 ( i.e. an interface symmetric under spin reversal ) .
the numerical data for the component ratios of the vacuum and the first two excited states is shown on the graphs below . at @xmath67 , @xmath199 . ] at @xmath67 , @xmath199 . ] at @xmath67 , @xmath199 . ] at @xmath67 , @xmath199 . ] at @xmath67 , @xmath199 . ] at @xmath67 , @xmath199 . ]
we see that the ratios for the vacuum and the first excited level each behave qualitatively very similar to the vacuum ratios for the massive flows discussed in section [ im_massive ] .
the ratios @xmath264 blow up near @xmath331 while @xmath329 blow up near @xmath332 . in each case
the ratios @xmath261 , @xmath333 remain finite and increasing functions past the blow up points .
the ratios @xmath334 while remaining finite do not show any tendency to level at a constant value . while we can not say with certainty that the oscillations between the blow up points will continue everything points towards the non - convergent oscillatory scenario .
in addition to this numerical indications we would like to remark that the alternative scenario in which we approach a conformal interface symmetric under spin reversal would be very hard to envisage from a general point of view .
clearly at large distances the magnetic field perturbation dominates and it breaks this symmetry .
in this paper we showed how pairings between states arising in rg interfaces of @xcite can be calculated numerically using truncated hamiltonian techniques . as illustrated by the exactly solvable case of the ising model with zero magnetic case in order to read off the numerical values of the pairings ( or rather their ratios ) one needs to go to very large values of dimensionless couplings . at these values
one can not a priory trust any results .
perturbative corrections are organised in ratios of couplings to the truncation energy and are very large in this region .
some aberrations in the low - lying spectrum that one can easily spot are the non - linearity of the vacuum energy dependence on scale and non - constance of the mass gap : @xmath335 . these can be easily spotted in the tcsa sample of data presented on the plots below .
the tffsa vacuum energy behaves much better even at very large scales @xmath181 .
the vacuum energy remains linear with a very good accuracy ( regression variance is about 0.01 ) .
the mass gap calculated using tffsa deviates from constant at big enough @xmath149 but the deviation is significantly smaller than in the tcsa data .
this means that in general we can not trust the energy eigenvalues and eigenvectors for such large values of scale , and yet our numerical results ( both tcsa and tffsa ) indicate that ratios of low lying components of the vacuum eigenvector are well behaved at large scales and the non - perturbative errors for them remain well bounded .
this suggests that although we need to go to very large scales to read off the asymptotic values , these quantities are more robust against truncation errors than the energy eigenvalues and eigenvectors as a whole .
at the moment we have no conceptual understanding of this fact which is crucial for the tcsa scheme to work for calculating the ir asymptotics of component ratios .
perturbative corrections also need to be understood as well with a suitable quantitative method for their incorporation needs to be worked out .
another important issue that needs further understanding is renormalisation of wave functionals . as discussed in the introduction in some cases additional boundary counter terms
may be needed to renormalise wave functionals in the interaction picture .
this may occur when vector fields are present in the ope of the perturbing operators . in the bulk perturbation such terms
do not need any counter terms due to rotational symmetry but from the point of view of the boundary they are scalars and may lead to additional divergences for collisions at the boundary ( where rotational symmetry is broken ) .
such counter terms are local along the boundary but what is their manifestation in the hamiltonian formalism ( and tcsa ) is not clear .
there is a good chance , in our opinion , that for superrenormalisable perturbations the above conceptual issues do not impede using the raw tcsa data for identifying rg interfaces and rg boundaries .
it would be interesting to explore systematically other examples such as the tricritical ising model or potts model .
for the tricritical ising model there are 4 relevant operators and 6 elementary conformal boundary conditions
. it would be interesting to find out how the three - dimensional space of massive flows breaks up into regions according to their rg boundaries .
also a conformal interface for the rg flow from the tricritical ising to the critical ising was put forward in @xcite .
it will be interesting to test that proposal numerically using tcsa .
we hope to report some answers to these questions in a near future @xcite .
the author is grateful to adam nahum , slava rychkov , cornelius schmidt - colinet , and g ' abor tak ' acs for stimulating discussions and comments on the manuscript .
he also thanks the organisers of the workshop `` conformal field theories and renormalisation group flows in dimensions @xmath336 '' at galileo galilei institute ( ggi ) in florence and ggi staff for hospitality .
this work was supported in part by stfc grant `` particle theory at the higgs centre '' , st / l000334/1 .
all numerical results presented in the paper were obtained using wolfram _ mathematica _
package v. 10.2 .
p. fonseca and a. zamolodchikov , _ ising field theory in a magnetic field : analytic properties of the free energy _ ,
journal of statistical physics , vol .
110 , issue 3 ( 2003 ) pp .
527 - 590 ; arxiv : hep - th/0112167 . | perturbing a cft by a relevant operator on a half space and letting the perturbation flow to the far infrared we obtain an rg interface between the uv and ir cfts .
if the ir cft is trivial we obtain an rg boundary condition .
the space of massive perturbations thus breaks up into regions labelled by conformal boundary conditions of the uv fixed point . for the 2d critical ising model perturbed by a generic relevant operator
we find the assignment of rg boundary conditions to all flows .
we use some analytic results but mostly rely on tcsa and tffsa numerical techniques .
we investigate real as well as imaginary values of the magnetic field and , in particular , the rg trajectory that ends at the yang - lee cft .
we argue that the rg interface in the latter case does not approach a single conformal interface but rather exhibits oscillatory non - convergent behaviour .
.5 in * rg boundaries and interfaces in ising field theory * + .250 in .5 in anatoly konechny 0.5 cm _ department of mathematics , heriot - watt university + edinburgh eh14 4as , united kingdom + and + maxwell institute for mathematical sciences + edinburgh , united kingdom + _ e - mail : [email protected] .5 in |
directional detection allows for unambiguous observation of dark matter ( dm ) even in presence of insidious backgrounds . when a weakly interacting massive particle ( wimp ) collides with a nucleus in the active mass of the detector , the direction of the nuclear recoil encodes the direction of the incident particle . for detectors consisting of a low - pressure gas ( about 50 torr ) ,
the typical length of such a recoil is 1 - 2 mm , which is sufficiently long to be reconstructed .
the simplest models of the distribution of wimps in our galaxy suggest that the orbital motion of the sun about the galactic center will cause an earth - bound observer to experience a wimp wind with speed 220 km / s ( the galacto - centric velocity of the sun ) , originating from the direction of the sun s motion . because the direction of this wind is inclined by 42@xmath1 with respect to the rotational axis of the earth , the average dm direction changes by almost 90@xmath1 every 12 hours @xcite . an ability to measure such a direction would allow for a powerful suppression of insidious backgrounds ( e.g. neutrons ) as well as a unique instrument to test local dm halo models .
our detector consists of a low - pressure tpc with optical readout .
the target gas is @xmath0 , whose spin 1/2 fluorine nuclei provide the ideal target material to detect spin - dependent interactions @xcite . in addition
, @xmath0 has high scintillation efficiency and low diffusion , and is ideal for underground detectors because it is non - toxic and non - flammable .
when an incoming wimp collides with a @xmath0 molecule at the design pressure of 50 torr , the emitted fluorine nucleus recoils 1 - 2 mm , ionizing the surrounding gas .
the primary electrons drift toward the amplification region , where charge multiplication takes place . to achieve 2d resolution ,
the amplification plane is built out of woven meshes @xcite . in the avalanche
, scintillation photons are produced together with the electrons in the ratio of 1:3 @xcite .
such photons are imaged by a ccd camera triggered by a pmt or electronic readout of the charge collected on the meshes .
the dmtpc detector is designed to measure the following quantities : * total light collected by the ccd , which is proportional to the energy of the nuclear recoil ; * length of the recoil track projected onto the amplification plane ; * length of the recoil track perpendicular to the amplification plane inferred from the width of the pmt signal ; * versus of the nuclear recoil ( `` head - tail '' ) as determined by the shape of the energy loss along the recoil track ( de / dx ) . at the typical energy of these recoils , de / dx decreases as the track slows down .
the simultaneous measurement of energy and length of the recoil effectively rejects electrons and @xmath2 tracks .
several prototypes ( figure 1 , top left ) have been built to prove the dmtpc detector concept .
alpha tracks from @xmath3am and low - energy neutrons from @xmath4cf are used to calibrate the device and measure its performance @xcite . for nuclear recoils of 100 kev , where the angular resolution is 15@xmath1 , we have achieved an energy resolution of @xmath5 10% .
typical gas gains are @xmath5 10@xmath6 - 10@xmath7 .
the intrinsic spatial resolution is of the order of 100 @xmath8 m , adequate to image recoils of 1 - 2 mm length with typical diffusion of 200 - 700 @xmath8 m .
the detector performance has been simulated using a combination of srim @xcite , casino @xcite , and geant4 @xcite .
the data - mc agreement is better than 10% .
a 10-liter detector @xcite has been built and is being commissioned in the laboratory .
underground operation is expected in early 2009 with the goal of studying backgrounds and place our first limits on spin - dependent interactions .
studies of the nuclear recoils induced by low - energy neutrons from @xmath4cf have demonstrated that the dmtpc detector can measure the energy , direction , and versus of recoiling nuclei @xcite .
figure 1 ( bottom left ) shows a typical nuclear recoil reconstructed in the dmtpc detector at a pressure of 75 torr .
the neutrons were traveling right to left .
the decreasing de / dx along the track direction is well visible , indicating the capability of determining the sense of the direction ( `` head - tail '' ) on an event - by - event basis .
mc studies indicate excellent head - tail discrimination can be obtained for nuclear recoils above 70 kev in cf@xmath9 at a pressure of 50 torr .
the dmtpc collaboration is designing a 1-m@xmath10 detector to be operated in an underground site .
this detector has two 1-m@xmath11 amplification planes . in the current design ,
each plane serves two 25-cm drift regions , and is imaged by 10 ccd cameras and 10 pmts . by running this device for one year at 100 torr
we will obtain an exposure of 100 kg - days . assuming a threshold of 50 kev and passive neutron shielding
, this device will allow us to set a limit on the spin - dependent cross section at @xmath12 , as shown in figure 1 ( right ) .
this high sensitivity is achieved despite the limited mass due to the excellent sensitivity that fluorine has to spin - dependent interactions @xcite and because directionality increases the sensitivity to dark matter by over one order of magnitude @xcite .
a larger detector with an active mass of a few hundred kg will explore a significant portion of the mssm parameter space .
the observation of directional wimp signal by this detector will allow us to test our understanding of the local dm halo model .
this detector is an ideal candidate for the dusel laboratory in south dakota .
this work is supported by the advanced detector research program of the u.s .
department of energy ( contract number 6916448 ) , the national science foundation , the reed award program , the ferry fund , the pappalardo fellowship program , the mit kavli institute for astrophysics and space research , and the mit physics department .
9 d. n. spergel , phys .
d37 , 1353 , ( 1988 ) .
d. dujmic et .
[ dmtpc collaboration ] , astroparticle physics 30 ( 2008 ) 58 - 64 .
a. kaboth et al .
[ dmtpc collaboration ] , arxiv:0803.2195[phy.ins - det ] .
d. dujmic et .
[ dmtpc collaboration ] , nim a 584:327 - 333 , ( 2008 ) .
j. f. ziegler et .
pergamon press , new york , 1985 .
the code is available online at www.srim.org .
d. drouin et al . , www.gel.usherb.ca/casino/what.html .
s. agostinelli et al . , methods in physics research a 506 ( 2003 ) 250 - 303 .
d. dujmic et .
[ dmtpc collaboration ] , these proceedings .
j. r. ellis and r. a. flores , phys .
b 263 ( 1991 ) 259 . a.green and b.morgan , astropart.phys . 27
( 2007 ) 142 - 149 .
m. s. alenazi , p. gondolo , phys . rev .
d 77 , 043532 ( 2008 ) . | directional detection of dark matter allows for unambiguous direct detection of wimps as well as discrimination between various dark matter models in our galaxy .
the dmtpc detector is a low - pressure tpc with optical readout designed for directional direct detection of wimps . by using @xmath0 gas as the active material
, the detector also has excellent sensitivity to spin - dependent interactions of dark matter on protons . |
many studies of globular cluster evolution have focused on internal mechanisms which drive evolution .
this work has produced a clear picture in which initial stellar evolution causes significant mass loss from a nascent cluster ( e.g. chernoff & weinberg 1990 ) ; two - body relaxation leads to mass segregation ( e.g. inagaki & saslaw 1985 ) and core collapse in surviving clusters ( e.g. cohn 1980 ) ; binary heating halts collapse ( e.g. lee et al . 1991 ) ; and the cluster continuously loses mass due to the escape of stars , eventually undergoing complete evaporation ( e.g. lee & goodman 1995 ) .
it is also recognized that the galaxy influences cluster evolution .
the time - dependent tidal field heats clusters and tidal limitation aids in the removal of escaping stars .
previous investigations have considered disk shocking , bulge shocking and tidal limitation , concluding that each will play a role , particularly in the inner galaxy ( e.g. spitzer & chevalier 1973 ; chernoff & shapiro 1987 ; aguilar , hut & ostriker 1988 ; weinberg 1994 ) .
in addition , recent observational studies showing correlations of cluster properties with galactocentric position indicate the measurable influence of the galaxy ( e.g chernoff & djorgovski 1989 ; djorgovski et al 1993 ; djorgovski & meylan 1994 ) . the principal tool used in studies of cluster evolution over the last decade - and - a - half has been direct solution of the fokker - planck equation ( cohn 1979 )
. however , most of these calculations have excluded external effects .
recently , using time - dependent perturbation theory to investigate disk shocking , weinberg ( 1994 ) demonstrated that resonant forcing can perturb internal stellar trajectories beyond the limit expected from adiabatic invariance .
this indicates that the galaxy plays a greater role in cluster evolution than previously thought and motivates new studies of cluster evolution which combine internal and external effects .
the importance of external heating requires us to re - examine the current picture of internally - driven evolution .
in particular , external effects will influence the collapse rates , evaporation times and general physical properties derived in previous calculations .
the present work compares this behavior with and without heating over a wide range of cluster properties to present a revised view .
this study also examines the survival and disruption characteristics of clusters on a range of galactic orbits to shed light on the initial conditions of the cluster system .
the results demonstrate that evolution does indeed depend strongly on position and orbit , further implying that observed cluster properties have been largely determined through dynamics .
our study rests on a linear theory of external heating based on weinberg s ( 1994 ) treatment of disk shocking which we include in numerical solutions of the fokker - planck equation .
nearly all previous work has emphasized impulsive shock heating due to a single passage through the disk or bulge .
the work presented here describes resonant heating due to the time - varying tidal field encountered on periodic orbits of the cluster in the galaxy- an effect we refer to as _ orbit _ heating . in this context
, shock heating is seen to result from the broad resonances caused by an impulsively applied external force .
although our treatment of external heating can include the influence of any component of the galactic potential , here we consider only the spheroid in order to allow precise definition of the physical behavior and preliminary description of the evolutionary trends .
the present study includes heating on cluster orbits in the isothermal sphere and is used to study cluster evolution from initial king model states to the point of complete evaporation on a range of orbits in the galaxy .
our conclusions , therefore , place only lower limits on the overall rate of cluster evolution but are significant nonetheless .
the plan of the paper is as follows .
we derive the linear theory of external heating in [ sec : eh ] and discuss its physical interpretation in [ sec : sumep ] . in [
sec : calc ] , the numerical implementation is described . in [ sec : results ] we present the results of our study of cluster evolution under the combined influence of internal and external effects . finally , in
[ sec : disc ] , we discuss the implications of the results for the milky way globulars .
readers concerned primarily with the effects of heating and its evolutionary consequences may skip [ sec : eh ] without loss of continuity .
the physics behind the perturbation theory discussed below can be summarized as follows .
each orbit in the cluster acts like a pendulum with two - degrees of freedom ( cf . binney & tremaine 1987 , chap .
3 ) . the time - dependent tidal field can drive the pendula at a discrete or continuous spectrum of frequencies depending on whether the perturbation is quasi - periodic or aperiodic , respectively . because the temporal variation discussed here is caused by the cluster s orbit in the spherical component of the galaxy ,
the spectrum is discrete . for disk shocking described by weinberg ( 1994 ) ,
the spectrum is continuous . in both cases ,
the energy of each orbit changes as it passes through the resonance .
the accumulated effect of all possible resonances on all orbits , drives the secular evolution of the equilbrium distribution function ( df ) .
the expressions given below are valid for both periodic and aperiodic cases .
we compute the evolution by expanding the boltzmann equation to first order and solving for the perturbed distribution function ( neglecting self - gravity ) .
the first - order change phase mixes but second order energy input leads to an induced phase space flux which helps drive cluster evolution .
n - body comparisons shown in appendix [ sec : compsim ] indicate that the self - gravity of the tidally - induced wake has negligible effect for cases of interest here .
we use a locally inertial reference frame which is centered on the cluster and has axes fixed in space ( see appendix [ sec : hamiltonian ] for derivation ) .
the unperturbed hamiltonian is therefore completely separable , implying the existence of action - angle variables .
this frame allows the internal dynamics to be defined in accordance with the standard fokker - planck technique ( e.g. cohn 1979 ) which uses an energy - space df @xmath1 and depends on the adiabatic invariance of the radial action . within this framework
, we derive a version of the formalism presented by weinberg ( 1994 ) which describes heating of globular clusters on arbitrary orbits in external potentials .
the linearized boltzmann equation is a linear partial differential equation in seven variables .
using action - angle variables , we can separate the equation and employ standard dfs constructed according to jeans theorem ( binney & tremaine 1987 ) .
the explicit form of the linearized boltzmann equation is @xmath2 where @xmath3 is the vector of angles , and @xmath4 are the conjugate actions .
the quantities @xmath5 and @xmath6 depend on the actions alone .
the small variation in galactic potential over a typical cluster size allows quadratic expansion of the tidal field ( see appendix [ sec : hamiltonian ] for details ) .
we may thus define @xmath7 and expand in a fourier series in action - angle variables ( e.g. tremaine & weinberg 1984 ) .
each term @xmath8 in the fourier series is the solution of the following differential equation : @xmath9 where @xmath10 and @xmath11 the inhomogeneous equation may be solved using a green s function ( e.g. birkhoff & rota 1962 , p.39 ) to give the time - dependence for each coefficient of the perturbed df @xmath12 where we have assumed that the perturbation begins at time @xmath13 .
the rate of change in energy arising from the perturbation follows from hamilton s equations ( weinberg 1994 ) .
the total phase - averaged change in energy can be written as @xmath14 substituting for @xmath8 from equation ( [ eq : f1l ] ) yields @xmath15 which represents the heat input due to the perturbation during an interval @xmath16 .
this expression is valid for finite - duration , aperiodic perturbations such as disk passage as well as periodic perturbations which arise on regular orbits in the galaxy . in particular , weinberg s ( 1994 ) results for disk shocking are obtained from equation ( [ eq : de ] ) by substituting the tidal amplitude appropriate to the disk profile for @xmath17 and integrating over the interval @xmath18 assuming a linear trajectory . for periodic perturbations
it is more suitable to derive the asymptotic heating rate ( e.g. landau & lifschitz 1965 , p.151 ) .
we first expand the tidal amplitude in a fourier series @xmath19 and substitute into equation ( [ eq : de ] ) . taking the limit @xmath20 and assuming the onset of the perturbation at @xmath21
, we obtain @xmath22 integrating @xmath23 over inclination and angular momentum , we obtain the change in energy which defines the induced change in the distribution , given by a one - dimensional continuity equation in energy space ( appropriate to the 1d fokker - planck formulation employed below ; see appendix [ sec : fluxeq ] for derivation ) : @xmath24 where @xmath25 is the phase space volume .
this is called the equation of quasilinear diffusion in the plasma literature ( e.g. stix 1992 ) .
the term _ quasilinear _ refers to the proportionality of the heating rate to the squared amplitudes of the linear modes .
the linear modes arise from the resonant forcing of stellar orbits by a periodic perturbation .
the competition between two - body relaxation and this externally induced phase space flux can strongly influence globular cluster evolution , as we will show below .
below we will need the heating rate for a cluster orbiting in the isothermal sphere . for most galaxies ,
the small variation in potential over a typical cluster size allows quadratic expansion of the tidal field .
therefore , the perturbing hamiltonian is : @xmath26 , \label{eq : perpot}\end{aligned}\ ] ] where @xmath27 is the angular rotation speed at the orbital radius at time t and @xmath28 is the instantaneous azimuthal angle of orbit . using equations ( 10 ) and ( 11 ) in weinberg ( 1994 ) , we can write the perturbation as a series in action - angle variables : @xmath29 where @xmath30 @xmath31 is the inclination of the orbital plane and @xmath32 is a rotation matrix ( e.g. tremaine & weinberg 1984 ) .
the angle @xmath33 is the difference between the mean azimuthal angle @xmath34 and the azimuthal angle in the orbital plane .
we substitute this expansion into equation ( [ eq : edot ] ) and average over inclination and angular momentum to derive the heating rate @xmath35\fsum |a_n|^2\delta(\ldomega - n\omega ) \nonumber\\ + \biggl[{1\over 30}\delta_{l_2 0}+{1\over 20}\delta_{l_2 |2|}\biggr ] \fsum |b_n|^2\delta(\ldomega - n\omega)\biggr\},\end{aligned}\ ] ] where @xmath36 @xmath37 and p is the period of the cluster orbit . for an isotropic df , @xmath38 .
we also explore the effect of anisotropy using merritt - osipkov models ( e.g. binney & tremaine 1987 ) .
the distribution function takes the form @xmath39 , where @xmath40 , @xmath41 .
the anisotropy increases with decreasing anisotropy radius , @xmath42 .
a cluster orbiting in the galaxy feels a time - dependent tidal field . a typical orbit is periodic and introduces a periodic external force on orbits of cluster stars . as described in
[ sec : eh ] , resonant heating occurs when the periods of stellar orbit and external force coincide , leading to repeated acceleration and increase in the energy of individual orbits . integrated over many periods , the energy gained by the orbit increases linearly with time ( c.f .
[ eq : res ] ) .
energy absorption eventually leads to the evolution of individual orbits .
( see appendix [ sec : imp ] for discussion and numerical implementation of finite duration resonances ) .
this in turn drives the secular evolution of the cluster potential .
orbits can either gain or lose energy to the tidal field depending on the particular resonance .
for example , in disk galaxies with flat rotation curves it is well - known that the inner lindblad resonance loses energy to a perturbation while an outer lindblad resonance gains energy .
however , for isotropic distribution functions with @xmath43 , the perturbation always heats the system on average though some localized regions of phase space may lose energy .
non - resonant interaction has no net effect on an orbit .
successive maxima in the external force tend to accelerate and decelerate the star equally , leading to asymptotic cancellation as long as the initial transients remain linear ( i.e. do not alter the intrinsic frequency of the star with an initial jolt ) . over short times ,
non - resonant heating does occur because the time duration is insufficient for complete cancellation to occur .
non - linear transient or _ impulsive _ heating leads to rapid change in orbital energies as a rapidly applied force ` kicks ' a star regardless of its orbital frequency .
however , the standard impulse approximation , when used to describe a periodic perturbation , ignores the long - term decay of transient energy in the linear limit as well as the linear growth in energy at the resonances . for most cases of interest ,
heating rates are in the linear limit , implying that external influence depends primarily on secular transfer of energy through orbital resonances . to illustrate the behavior of transients and transient decay , figure [ fig : k5.in.3.comp ] compares the exact time - dependent energy input given by equation ( [ eq : de ] ) with the energy input defined by the asymptotic heating rate equation ( [ eq : res ] ) .
transients decay rapidly at low energy and more slowly at high energy .
empirically , we find that two to three galactic orbital periods are required before the asymptotic limit is effectively reached .
this treatment therefore adequately describes all but the outermost halo clusters for which initial transients may still be important .
the comparisons of perturbation theory with n - body simulation shown in appendix [ sec : compsim ] demonstrate the validity of the approach .
the magnitude of the heating rate is determined by the cluster profile , density and orbit .
the profile and density define the distribution of internal orbital frequencies and the cluster orbit defines the external forcing frequencies and amplitude . for a cluster of fixed profile and mass , the density is determined by the tidal radius .
individual clusters may not be tidally limited due to initial conditions or heating - driven expansion .
therefore we use the function @xmath44 to parameterize the fraction of the total cluster mass enclosed within the instantaneous pericentric inner lagrange point , @xmath45 .
this function depends on the profile and the ratio of cluster mean density to the mean density required by tidal limitation .
a tidally - limited cluster has limiting radius , @xmath46 , and therefore @xmath47 , while a tidally - unlimited has @xmath48 and therefore @xmath49 .
heating rates for a given orbit increase as @xmath44 decreases .
the perturbing potential in the logarithmic sphere , equation ( [ eq : perpot ] ) , heats clusters on orbits of any eccentricity . the tide transfers energy and angular momentum to the cluster through the resonances , which unbinds stars .
on circular orbits , the tidal field creates a triaxial perturbation of constant amplitude proportional to @xmath50 rotating with fixed pattern speed @xmath51 . on eccentric orbits ,
conservation of center - of - mass angular momentum introduces time - dependent amplitude and rotation rate .
this produces more resonances .
tidal torquing can also induce a net spin . the rate of external heating is also influenced by our choice of equilibrium phase space distributions .
for example , according to jeans theorem , one can define equilibria in the rotating frame of a circular cluster orbit using the jacobi constant , @xmath52 ( e.g. heggie & ramamani 1994 ) . by transforming to the frame
in which the perturbation is time - independent , we remove the resonances from the problem .
we can therefore choose a bound distribution of orbits in @xmath52 using the limiting zero - velocity surface , so no stars are lost and the cluster experiences no net tidal heating , although inertial energies and angular momenta are not conserved . using @xmath1 instead of
@xmath53 leads to heating in the analogous case because we can not choose orbits which are strictly bound . in any case
, a real cluster can not reach true equilibrium because it is bound and therefore undergoes relaxation .
in fact , as is shown below , it is typically a competition between external heating and relaxation due to strong resonances with diffused core stars that strongly influences cluster evolution .
to determine the influence of external heating on cluster evolution , we conduct a series of fokker - planck calculations which begin with king model initial conditions and run through core collapse to complete evaporation .
relaxation is computed using the multi - mass code of chernoff & weinberg ( 1990 ) which solves henon s ( 1961 ) orbit - averaged fokker - planck equation .
core heating is included in the form described by lee et al ( 1991 ) with a time step that supresses stochastic core osillations .
implementation of external heating is detailed in appendix [ sec : imp ] .
the comparisons shown in appendix [ sec : compsim ] are used to test the implementation .
lcccccccc & 8parameters + + process&@xmath54&@xmath55&@xmath56&@xmath57&@xmath44&@xmath58&@xmath59&@xmath60 + + relaxation&@xmath61&@xmath61&@xmath61&@xmath61&@xmath61&@xmath61 & & + + external heating&&&&@xmath61&@xmath61&@xmath61&@xmath61&@xmath61 + + core heating&@xmath61&@xmath61&@xmath61&@xmath61&@xmath61&@xmath61 & & + each physical process depends on the input model parameters listed in table [ tab : param ] .
the total mass is denoted by @xmath58 and the concentration by @xmath57 .
orbits in the isothermal sphere are defined by their energy @xmath59 and angular momentum @xmath62 . in place of absolute angular momentum @xmath62
, we use the relative angular momentum @xmath63 , where @xmath64 is the angular momentum of a circular orbit with energy @xmath59 .
the value @xmath65 denotes a radial orbit and the value @xmath66 denotes a circular orbit .
the apocentric , pericentric and mean orbital radii are denoted @xmath67 , respectively .
we represent the galactic potential as a singular isothermal sphere with rotation velocity @xmath68 km / s .
we consider a range of initial values for @xmath44 .
if the young , rich lmc clusters are representative of young globular clusters , @xmath44 may be significantly smaller than unity initially ( elson , fall & freeman 1987 ) .
furthermore , as discussed in
[ sec : epc ] , formation scenarios can imply varying degrees of tidal truncation for an individual cluster depending on the local conditions under which it forms and the orbit on which it travels .
the distribution of stellar masses in the cluster is given by a power - law mass spectrum , @xmath69 , with upper and lower mass limits @xmath55 and @xmath56 , respectively .
fiducial values @xmath70 , @xmath71 and @xmath72 are adopted in [ sec : ebs2 ] to represent the cluster mass spectrum following the period of strong stellar evolution when relaxation , tidal heating and binary heating dominate cluster evolution .
the importance of stellar evolution diminishes after @xmath73 for @xmath70 and @xmath71 which corresponds to the main sequence lifetime for a @xmath74 a - star .
the effect of changing the mass spectrum is explored in [ sec : ims ] .
because heating rates depend on cluster profile , tidal truncation and orbit , comparisons in different physical regimes are needed to demonstrate the primary influences of heating on cluster evolution .
we choose four specific examples listed in table [ tab : examples ] .
example 1 compares the relative strengths of heating rates on different orbits using physically identical clusters , each of which is tidally - limited at its orbital pericenter . in this case , because the average tidal strength is largest on circular orbits , heating is also strongest on circular orbits and decreases with eccentricity ( figure [ fig : der.comp ] ) . to investigate the effect of heating on long - term evolution , we compare evaporation times , @xmath75 , for tidally limited clusters of different mass , concentration and @xmath60 .
table [ tab : t_ev ] shows @xmath75 normalized by the circular , @xmath0 @xmath76 case ( arbitrary scaling to physical units is provided in
[ sec : eoc ] ) . in these comparisons ,
clusters of a particular mass and concentration are identical and clusters of differing mass and concentration possess the mean density required for tidal limitation .
for identical clusters , @xmath75 decreases monotonically with @xmath60 , reaching a minimum for circular orbits .
evaporation times can decrease by a factor of two in circular cases when tidal heating is included .
the relative evaporation times reflect the relative strength of heating rates shown in figure [ fig : der.comp ] .
heating accelerates evolution because external forcing efficiently torques and expels high - energy core stars on radial orbits , as noted by oh & lin ( 1992 ) in n - body calculations .
this reduces the local relaxation time in the core , enhances relaxation rates and causes rapid evaporation .
spitzer & chevalier ( 1973 ) noted this effect in certain regimes of disk shocking , interpreting it as an increase in the core - halo temperature gradient ( see also chernoff & shapiro 1987 , weinberg 1994 ) .
for the highest eccentricities , @xmath75 is only slightly shorter than with no heating , demonstrating the insignificance of high - eccentricity heating in tidally - limited clusters . in many cases ,
evaporation time does not vary strongly with concentration for the same orbit and mass , indicating that overall mass loss rates are insensitive to initial concentration . in the exceptional @xmath77 and @xmath78 , @xmath79 , @xmath80 cases , heating causes rapid disruption because these clusters have low binding energy and long relaxation times and are easily torn apart by the tide . [
cols="<,^,^,^,^,^,^,^,^",options="header " , ] while example 1 compares heating rates as a function of eccentricity in identical clusters , the orbits occupy different regions of the galaxy ( c.f .
table [ tab : examples ] ) .
in example 2 , we consider clusters in similar regions by comparing tidally - limited clusters on orbits of equal mean radius . because they are tidally truncated , these clusters still undergo the same rate of heating relative to internal energies shown in figure [ fig : der.comp ] . however , cluster densities vary due to differences in orbital angular frequencies . for fixed cluster mass
, this implies that tidal radii will vary . the tidal radius @xmath81 decreases with the increased perigalactic angular frequency at higher eccentricity .
this increases the mean density and decreases the dynamical time @xmath82 , producing shorter relaxation times , larger evaporation rates and , as a result , shorter lifetimes as compared to example 1 .
a cluster with @xmath83 in example 2 will evaporate in 1/7 the time of a cluster with @xmath83 in example 1 and 1/5 the time of a cluster with @xmath84 ( the circular case ) .
lccccc @xmath79&&&@xmath60 & & + + @xmath58 ( @xmath85)&1.0&0.9&0.6&0.3&nh@xmath86 + @xmath87&0.65&0.95&1.27&1.29&1.30 + @xmath88&0.94&3.69&9.61&10.9&11.2 + + @xmath76&&&@xmath60 & & + + @xmath58 ( @xmath85)&1.0&0.9&0.6&0.3&nh + @xmath87&1.00&1.15&1.34&1.37&1.38 + @xmath89&3.20&4.25&5.64&6.10&6.20 + @xmath88&5.20&7.01&10.31&11.43&11.82 + + @xmath90&&&@xmath60 & & + + @xmath58 ( @xmath85)&1.0&0.9&0.6&0.3&nh + @xmath87&1.12&1.27&1.40&1.41&1.42 + @xmath89&3.97&4.70&5.95&6.25&6.37 + @xmath88&6.35&8.29&11.06&11.96&12.13 + 6 since cluster orbits are generally unknown , the degree of tidal truncation at pericenter can not be directly inferred .
so , in example 3 , we assume that an observed cluster lies at its average orbital radius for a range of eccentricity and is tidally limited for zero eccentricity .
the mass within the pericentric inner lagrange point @xmath44 can be substantially less than unity on eccentric orbits ( table [ tab : examples ] ) .
this leads to stronger heating than found on the same orbits in example 1 ( see figure [ fig : case3 ] ) .
for @xmath91 the heating rate is much larger than in example 1 even though only small amounts of mass overlie @xmath45 . for @xmath83 strong impulsive heating or bulge shocking (
e.g. aguilar et al 1988 ) occurs due to the increase in tidal amplitude .
example 4 shows the dependence of heating rates on degree of tidal truncation for a fixed @xmath83 orbit .
figure [ fig : case4 ] illustrates the dependence of heating on both @xmath60 and @xmath44 : significant heating will occur on orbital timescales for @xmath83 and @xmath92 .
strong heating for @xmath93 will also occur because these orbits have larger heating rates for the same value of @xmath44 .
table [ tab : examples ] shows the variation in cluster size and dynamical time with tidal truncation , indicating the corresponding variation in mean density .
the evolutionary consequences of the heating rates in example 4 are shown in table [ tab : bs_ev ] .
clusters of different mass and concentration have equal @xmath44 on the same orbit .
weakly bound clusters disrupt more easily because the resonances occur more deeply within the system .
survival of @xmath79 clusters decreases strongly with @xmath44 .
conversely , for small reduction in @xmath44 , survival of @xmath76 and @xmath90 clusters is enhanced as increased heating is offset by diminished relaxation . for @xmath76 ,
maximum enhancement occurs at @xmath94 . for @xmath90 , higher binding energies
lead to even longer lifetimes for more severe truncations .
further reductions in @xmath44 eventually lead to rapid disruption due to strong tidal shocking .
these results define a rough criterion for bulge shocking : for @xmath95 and @xmath93 , bulge shocking will occur for @xmath92 .
disruption for fixed @xmath44 and @xmath60 also implies disruption for larger @xmath60 because heating rates increase with @xmath60 . for @xmath96
, bulge shocking requires even smaller @xmath44 to cause disruption .
lccccc + @xmath79&&&@xmath44 & & + + @xmath58 ( @xmath85)&1.0&0.95&0.9&0.8&0.7 + @xmath87&1.3&1.3&1.1&0.4&0.2 + @xmath88&10.9&2.4&0.8&0.4&0.4 + + @xmath76&&&@xmath44 & & + + @xmath58 ( @xmath85)&1.0&0.95&0.9&0.8&0.7 + @xmath87&1.4&2.4&2.1&1.0&0.5 + @xmath88&11.4&12.1&5.2&1.5&0.5 + + @xmath90&&&@xmath44 & & + + @xmath58 ( @xmath85)&1.0&0.95&0.9&0.8&0.7 + @xmath87&1.4&2.7&2.4&2.3&2.0 + @xmath88&12.0&16.0&23.4&13.8&7.5 + + this series of comparisons establishes three important aspects of tidal effects on different orbits in a spherical potential : 1 ) low - eccentricity and circular orbit heating for tidally - limited clusters strongly accelerate evolution ; 2 ) high - eccentricity heating has little effect in tidally - limited cases but the high mean density found for typical orbital radii in the galaxy leads to short relaxation and evaporation times ; 3 ) high - eccentricity heating , or bulge shocking , becomes important when clusters are tidally - unlimited , although the exact effect depends on @xmath44 , @xmath60 , @xmath57 and @xmath58 . finally , an important consequence of strong tidal heating is suppression of the gravothermal instability .
although this may cause expansion and disruption , relaxation slows the expansion and can still produce mass segregation ( figure [ fig : bs.a_r ] ) .
therefore , _ mass segregation does not necessarily imply core collapse , a possibility that does not arise when neglecting external heating ( e.g. chernoff & weinberg 1990 ; drukier , richer & fahlman 1992 ) . observed clusters with king profiles and strong mass segregation ( such as m71 )
may reflect the influence of strong tidal effects
. _ the mass spectrum controls the rate of relaxation and interplay with external heating .
clusters with steep mass spectra or a narrow range of low mass stars have lower relaxation rates than do clusters with shallow mass spectra or a wide range in stellar mass ( e.g. chernoff & weinberg 1990 ) . here
we examine the competition between external heating and relaxation over a range in @xmath54 and @xmath56 in unheated and circularly heated tidally - limited clusters .
circular heating reduces evaporation times over a range in @xmath54 ( table [ tab : t_cc.ev ] ) .
roughly a factor of three reduction can occur for @xmath97 .
differences between heated and unheated clusters increase with @xmath54 because the slower relaxation rates at high @xmath54 are more readily enhanced . in addition , for fixed mass and concentration , heating reduces differences in evaporation time which depend on @xmath54 .
lcccl + @xmath76&&@xmath54 & + + @xmath58 ( @xmath85)&1.35&2.35&3.35 & + 5@xmath75 + @xmath89&2.7&3.2&5.3&heated + & 4.4&6.3&14.1&unheated + @xmath88&4.3&5.2&8.2&heated + & 8.6&12.3&25.9&unheated + 5@xmath98 + @xmath89&1.6&1.6&2.1&heated + & 2.2&1.9&2.3&unheated + @xmath88&2.7&2.7&3.6&heated + & 4.0&3.1&4.4&unheated + + @xmath90&&@xmath54 & + + @xmath58 ( @xmath85)&1.35&2.35&3.35 & + 5@xmath75 + @xmath89&2.8&4.0&7.4&heated + & 4.3&6.1&14.0&unheated + @xmath88&4.9&6.4&11.8&heated + & 8.1&12.1&26.8&unheated + 5@xmath98 + @xmath89&0.56&0.37&0.41&heated + & 0.59&0.37&0.38&unheated + @xmath88&1.06&0.71&0.76&heated + & 1.06&0.71&0.76&unheated + 4 @xmath99 ,
@xmath100 + heating also reduces core collapse times @xmath98 up to 33% ( table [ tab : t_cc.ev ] ) and masses remaining at core collapse up to a factor of two ( table [ tab : m_cc ] ) .
high concentration clusters maintain the same core collapse times in all cases but show decreased mass at core collapse .
the non - monotonic behavior of core collapse time with spectral index was also found by inagaki ( 1985 table ii ) in plummer law initial profile and chernoff & weinberg ( 1990 table 4 ) in king models .
this indicates a complex relation between concentration , mass segregation and core collapse .
heating supresses this behavior for @xmath76 .
lcccl @xmath76&&@xmath54 & + + @xmath58 ( @xmath85)&1.35&2.35&3.35 & + @xmath89&0.39&0.54&0.59&heated + & 0.65&0.88&0.93&unheated + @xmath88&0.30&0.43&0.50&heated + & 0.65&0.88&0.93&unheated + + @xmath90&&@xmath54 & + + @xmath58 ( @xmath85)&1.35&2.35&3.35 & + @xmath89&0.82&0.90&0.92&heated + & 0.93&0.98&0.99&unheated + @xmath88&0.78&0.87&0.90&heated + & 0.93&0.98&0.99&unheated + 4 @xmath99 ,
@xmath100 + evaporation times decrease with increasing @xmath56 ( figure [ fig : ml.rho_comp ] ) .
the decrease in @xmath75 with increasing mass range results from enhanced relaxation caused by a more extreme mass segregation instability .
a 25% range in the duration of strong stellar evolution for @xmath70 gives a range in mass limits of @xmath101 and results in very small differences in evaporation time .
another internal property that determines the influence of external heating is the anisotropy of the stellar orbit distribution .
figure [ fig : der.ani ] shows the variation of heating rate with anisotropy radius within a cluster .
heating increases with anisotropy due to efficient impulsive heating of radial orbits at apocenter .
however , heating rates for @xmath102 unbind orbits with @xmath103 in one orbital period @xmath104 and quickly alter the df .
the relaxation time is roughly 100 crossing times , so diffusion can not maintain the assumed level of radial anisotropy in the cluster halo .
we estimate that anisotropy radii of @xmath105 are sustainable through relaxation .
the interplay between heating and anisotropy seen here provides strong incentive to study the evolution of fully anisotropic dfs .
the dimensionless evaporation times for tidally - limited clusters discussed in
[ sec : ebs2 ] may be scaled to physical units using the following relation @xmath106 where @xmath107 is the orbital period at the tidal radius @xmath108 and @xmath109 the quantity @xmath110 is the effective perigalactic angular frequency of an orbit of given @xmath60 and apocentric radius @xmath111 due to tidal strain and centrifugal force ( defined in appendix [ sec : pilp ] ) . as an example , the dimensionless evaporation times given in table [ tab : t_ev ] are scaled to a range of apogalactica in figure [ fig : t_ev.scale ] .
clusters evaporate over a wide range of galactocentric radii depending on @xmath60 . in 10 gyr ,
clusters on circular orbits evaporate within 3 kpc , while those on @xmath83 orbits evaporate out to average radii of 15 kpc . here
we present a simple evolutionary scenario in which clusters form at apocenter with a range of mean density parameterized by @xmath112 , the fall - rees ( 1985 ) critical cloud density at galactocentric radius @xmath113 .
this parameterization is chosen to allow normalization with respect to a particular model .
other models ( e.g. harris & pudritz 1994 ; murray & lin 1992 ) can be similarly evaluated given expressions for initial protocloud densities as a function of galactocentic radius .
a range of density is used to define a range of @xmath44 for clusters at each radius , thereby illustrating characteristics which are independent of any particular model .
we only consider relaxation , external heating and binary heating although gas removal and stellar evolution will play an important role following formation .
these effects should weaken the potential and increase disruptive tendencies described here . in the first case , clusters form on eccentric @xmath83 orbits ( e.g. eggen , lynden - bell and sandage 1962 ) .
figure [ fig : bs.3 ] shows the resulting pattern of survival , disruption and evaporation for @xmath114 and @xmath80 clusters after 10 gyr .
clusters initially with @xmath0 do not surive within @xmath115 , reflecting the evaporation times shown above .
lower density clusters suffer disruption to even larger distances .
high mass clusters with @xmath116 can suffer disruption but none can evaporate .
cluster formation on less eccentric @xmath91 orbits shows the same qualitative pattern of survival , evaporation and disruption as above ( figure [ fig : bs.7 ] ) .
the consequences are less severe because the density contrast between formation at apocenter and tidal limitation at pericenter is not as great . in this case ,
low mass cluster survival is limited to regions beyond 5 kpc for clusters which are nearly tidally - limited .
the calculations presented above bear on our understanding of the observed mass and space distributions of galactic globular clusters .
we summarize some relevant properties for reference . in the djorgovski ( 1993 )
compilation , 65% of 130 clusters with distance estimates lie within the solar circle .
the overall peak of the luminosity function of galactic globulars is @xmath117 ( harris 1991 ) corresponding to a mass of @xmath118 ( where @xmath119 ) . the luminosity function varies little in this inner region .
our results imply that the observed characteristics of this inner population have evolved with time .
because @xmath0 clusters evaporate or lose large amounts of mass in a hubble time in the inner galaxy , clusters at the peak of the luminosity function have evolved from higher mass .
for example , at 6 kpc clusters on circular orbits with @xmath117 will evolve to @xmath120 in 10 gyr , losing roughly 40% of their initial mass .
inside the solar circle , clusters near the present peak had at least 30% more mass , depending on the orbit .
many clusters will also have vanished .
we predict that evaporation and disruption of @xmath0 clusters occur within @xmath121 for @xmath84 and within apocentric radii @xmath122 for @xmath83 . for intermediate @xmath60 , the destruction region
is bracketed by these limiting cases .
these results buttress arguments based on two - body relaxation times that the shape of the luminosity function stems from evaporation and disruption of a larger initial population of low mass clusters ( e.g. larson 1996 , okazaki & tosa 1995 ) .
the dependence of survival on orbit implies that the kinematic distribution of clusters has evolved as well .
clusters on high eccentricity orbits in the inner galaxy are least likely to survive due to both evaporation and bulge shocking .
this suggests a decrease in the net velocity dispersion for the rotating system of metal - rich and inner halo metal - poor clusters ( zinn 1993 ) .
this tendency may also partially account for observed differences between the kinematics of halo field stars and metal - poor globular clusters ( e.g. aguilar et al .
1988 ) .
finally , survival also depends strongly on initial cluster densities .
destruction is more pronounced for clusters with low initial density and low initial concentration due to bulge shocking .
bulge shocking can disrupt massive @xmath80 clusters on eccentric orbits out to 40 kpc provided @xmath123
. however , a proper assessment of the initial distribution of cluster densities requires cosmogonical considerations .
we conclude that the segment of the cluster population which is native to the milky way or which was accreted at an early time represents a dynamically selected sample , with current masses , orbits and densities all favored for survival over a hubble time of evolution .
tidal interaction with the galactic disk will amplify these effects .
details will be described in a subsequent paper .
the key conclusions are as follows : 1 .
time - dependent heating on low - eccentricity orbits accelerates evolution and sharply reduces evaporation times .
tidally limited clusters on high - eccentricity orbits have high internal density , leading to short evaporation times even though heating rates are negligible .
bulge shocking on high - eccentricity orbits can rapidly disrupt clusters over a wide range in mass and apogalactic radius when their densities are roughly a factor of 10 below the mean density required for tidal limitation .
evaporation and disruption have shaped the mass , orbit and density distribution of clusters .
in particular , clusters at the peak of the luminosity function had at least @xmath124 more mass depending on orbit .
evaporation on high - eccentricity orbits has decreased the velocity dispersion in the cluster kinematic distribution .
secondary results are as follows : 1 .
evaporation times do not strongly depend on concentration in most cases .
however , heating can lead to rapid disruption in massive clusters with low concentration because of the low binding energy and long relaxation time .
clusters disrupting due to heating may still show signs of mass segregation due to continued relaxation .
3 . heating accelerates evolution over a range of mass spectral index and reduces the dependence of evaporation time on different initial mass spectra .
4 . the development of anisotropy through relaxation in the core will increase evolutionary rates found in the isotropic distributions investigated here .
this work was supported in part by nasa award nagw-2224 .
aguilar , l. , hut , p. & ostriker , j. p. 1988
, apj , 335 , 720 binney , j. j. & tremaine , s. d. 1987 , galactic dynamics , ( princeton : princeton u. press ) bracewell , r. n. 1986 , the fourier transform and its applications , ( new york : mcgraw hill ) chandrasekhar , s. 1942 , principles of stellar dynamics , ( chicago : univ .
chicago press ) chernoff , d. f. & djorgovski , s. g. 1989 , apj , 339 , 904 chernoff , d. f. , kochanek , c. s. & shapiro , s. l. 1986 , apj , 309 , 183 chernoff , d. f. & shapiro , s. l. 1987 , apj , 322 , 113 chernoff , d. f. & weinberg , m. d. 1990 , apj , 351 , 121 cohn , h. n. 1979 , apj , 234 , 1036 cohn , h. n. 1980 , apj , 242 , 765 cohn , h. n. 1985 , in dynamics of star clusters , eds .
goodman , j. , and hut , p. , ( boston : d. reidel publishing ) djorgovski , s. , piotto , g. & capaciolli , m. 1993 , aj , 105 , 2148 djorgovski , s. , & meylan , g. 1994 , aj , 108 , 1292 drukier , g. a. , fahlman , g.g . &
richer , h. b. 1992 , apj , 386 , 106 eggen , o. j. , lynden - bell , d. & sandage , a. 1962 , apj , 136 , 748 elson , r. a. w. , fall , s. m. & freeman , k. c. 1987 , apj , 323 , 54 elson , r. , hut , p. & inagaki , s. 1987 , araa , 125 , 565 fall , s. & rees , m. 1985 , apj , 298 , 18 goldstein , h. 1980 , classical mechanics , ( reading , mass : addison - wesley ) , p.383 harris , w. e. 1991 , araa , 29 , 543 harris , w. e. & pudritz , r. e. 1994 , apj , 429 , 177 heggie , d. c. , & ramamani , n. , 1995 , mnras , 272 , 317 henon , m. , 1961 , ann .
dastrophys . , 24 , 369 hernquist , l. & ostriker , j. p. 1992 , apj , 386 , 375 hesser , j. e. , 1993 , in the globular - cluster galaxy connection , eds . smith , g. h. & brodie , j. p. , ( astr . soc .
pacific : san francisco ) inagaki , s. , in dynamics of star clusters , eds .
goodman , j. , and hut , p. , ( boston : d. reidel publishing ) inagaki , s & saslaw , w. 1985 , apj , 292 , 339 king , i. r. , 1962 , aj , 67 , 471 krall , n. a. , & trivelpiece , a. w. 1973 , principles of plasma physics , ( new york : mcgraw - hill ) landau , l. d. , & lifschitz , e. m. 1965 , quantum mechanics , non - relativistic theory , ( london : pergamon press ) larson , r. b. 1990 , pasp , 102 , 709 larson , r. b. 1994 , preprint lee , h. m. & goodman , j. 1995 , apj , 443 , 109 lee , h. m. & ostriker , j. p. 1987 , apj , 322 , 123 lee , h. m. , fahlman , g. g. , & richer , h. b. 1991 , apj , 366 , 455 lichtenberg , a. , & lieberman , m. 1983 , regular and stochastic motion , ( new york : springer - verlag ) lightman , a. & shapiro , s. 1978 , revs .
phys . , 50 , 437 lynden - bell , d. & kalnajs , a. 1972 , mnras , 157 , 1 lynden - bell , d. & wood , r. 1968 , mnras , 138 , 495 murray , s. d. & lin , d. n. c. 1992 , apj , 400 , 265 oh , k. s. & lin , d. n. c. 1992 , apj , 386 , 519 okazaki , t. & tosa , m. 1995 , mnras , 274 , 48 ostriker , j. p.1985 , in dynamics of star clusters , eds .
goodman , j. , and hut , p. , ( boston : d. reidel publishing ) ostriker , j. p. , spitzer , l. , jr .
& chevalier , r. 1972 , apjl , 176 , l51 richer , h. b. , & fahlman , g. g. 1989 , apj , 339 , 178 rosenbluth , m. n. , macdonald , w. m. & judd , d. l. 1957 , physrev , 107 , 1 spitzer , l. , jr .
1987 , dynamical evolution of globular clusters , ( princeton : princeton univ .
press ) spitzer , l. , jr . &
hart , m. 1971 , apj , 164 , 399 spitzer , l. , jr .
& chevalier , r. a. 1973 , apj , 183 , 565 stix , t. h. 1992 , waves in plasmas , ( new york : aip press ) szebehely , v. 1967 , theory of orbits , ( new york : academic press ) titchmarsh , e. c. 1986 , introduction to the theory of fourier integrals , ( new york : chelsea publishing company ) tremaine , s. d. & weinberg , m. w. 1984 , mnras , 209 , 729 weinberg , m. d. 1994 , aj , 108 , 1414 zinn , r. , 1993 , in the globular - cluster galaxy connection , eds . smith , g. h. & brodie , j. p. , ( astr . soc .
pacific : san francisco )
in the inertial galactocentric frame , the coordinate components of a cluster star are @xmath125 and its velocity components are @xmath126 where @xmath127 and @xmath128 are the coordinates and velocities of the member star measured relative to the cluster center of mass and @xmath129 and @xmath130 are the center - of - mass position and velocity of the cluster .
the hamiltonian for an individual star in these coordinates is therefore @xmath131 we introduce coordinates centered on the cluster with axes fixed in space through a canonical transformation using a generating function of the second kind ( goldstein 1985 ) .
this function can be written @xmath132 where @xmath133 is an arbitrary function of time .
the transformation obeys the conditions @xmath134 and @xmath135 , thus satisfying hamilton s principle .
the new hamiltonian @xmath136 so that ( assuming the summation convention throughout ) @xmath137 expanding the galactic tidal potential about the center of mass , we obtain @xmath138\nonumber\\ + \bigl[-{1\over 2}|\vec{v}_{com}|^2+\phi_g(|\vec{r}_{com}|)\bigr ] + { \partial f\over \partial t}.\end{aligned}\ ] ] the term in the second pair of brackets is an arbitrary function of time which arises as an ambiguity in canonical transformations ( goldstein 1985 ) .
we note that it equals @xmath139 , the negative lagrangian of the center - of - mass motion and that it can be eliminated by an appropriate choice of @xmath140 . in this case , setting @xmath141 ( the action associated with the center - of - mass motion ) gives the desired form of the hamiltonian for a star in the cluster frame : @xmath142.\ ] ] in the expansion of the tidal potential , we can ignore all but the lowest - order term because successive terms are proportional to @xmath143 ( @xmath144 ) relative to the second order term and , therefore , fall off quickly due to the small size of a cluster compared to the size scale of the galaxy .
we are thus left with the quadratic approximation to the galactic tidal field .
the expression for the perturbing potential , equation ( 8) , is obtained by evaluating this term of the expansion for the specific case of the logarithmic sphere .
to obtain an expression for the pericentric inner lagrange point , it is convenient to first transform to rotating coordinates having one axis aligned with the galactocentric radius of the cluster on its orbit .
we omit the details of the transformation here and simply give the expression for the effective potential at pericenter in the rotating frame : @xmath145-{1\over 2}|\vec{\omega_p}\times\vec{r}|^2.\ ] ] the first term is the cluster potential .
the second term is the quadratic tidal potential for the logarithmic sphere transformed to a rotating coordinate system .
the last term is the centrifugal potential arising from the angular frequency of rotation at pericenter .
the quantity @xmath146 is the pericentric angular frequency of the cluster orbit while @xmath147 is the angular frequency of a circular orbit at the pericentric radius which defines the tidal strain .
the pericentric inner lagrange point @xmath45 occurs at the instantaneous inflection point in the effective potential which lies along the galactocentric radius of the cluster .
we derive an expression for @xmath45 by considering the balance of forces which is implied by the effective potential .
taking the gradient , and considering the instantaneous point of equilibrium along the galactocentric radius gives @xmath148 where the left - hand side gives the cluster force while the right - hand side gives the centrifugal force and tidal strain , respectively .
to derive an expression for the inner lagrange point in terms of @xmath149 , we first rewrite the angular frequencies @xmath146 and @xmath147 in terms of the angular frequency of a circular orbit at apocenter @xmath150 . using conservation of angular momentum between apocenter and pericenter
gives @xmath151 where @xmath152 denotes the ratio of the angular frequency of the orbit to the angular frequency of a circular orbit with the same apocenter .
the flat rotation curve defines @xmath153 and substituting into equation ( [ eq : fb ] ) and solving for @xmath45 , we obtain @xmath154\biggl\}^{-1}.\ ] ] now we define the effective pericentric angular frequency @xmath155\biggl\}\ ] ] where @xmath156 for the logarithmic sphere .
the flux equation , equation ( [ eq : qld ] ) , can be derived from the @xmath157 formalism given in weinberg ( 1994 ) .
the function @xmath157 defines the externally induced change in the distribution function .
the general form for @xmath157 is @xmath158 where we have performed the phase - averaging to derive the perturbation as a function of the actions .
the quantity @xmath159 is a scalar function of the actions @xmath160 where @xmath161 is a fourier coefficient given by @xmath162 and other quantities are as defined in [ sec : eh ] . to derive the flux equation we note that @xmath157 can be written as a divergence : @xmath163 where @xmath164 .
this equation makes number conservation manifest in action space . to implement this term in a 1-dimensional fokker - planck scheme
, we must change variables from actions to @xmath165 and average over @xmath60 and @xmath166 to obtain the one - dimensional flux in energy space .
the transformation can be performed easily using the covariant form of the equation ( e.g. rosenbluth et al 1957 ) .
the divergence written in covariant form is @xmath167 where @xmath168 is a contravariant vector and @xmath169 is the determinant of the metric tensor , equal to the square of the jacobian . using this equation
, we transform to the new @xmath170 coordinates , which we denote using primes . transforming the contravariant vector @xmath171
gives @xmath172 where the quantity @xmath173 is the function defined above but now written in terms of the new variables .
@xmath174 is the function equivalent to @xmath171 in the new coordinates . noting that the jacobian is @xmath175
we may write @xmath157 in the new coordinates : @xmath176 averaging over @xmath177 @xmath178 gives the total change as a function of energy : @xmath179 where the fluxes in the @xmath180 and @xmath181 directions vanish due to the averaging .
fully expressed , the equation reads @xmath182\biggr\ } \biggl\{\int d\kappa\kappa j_{max}^2/\omega_1\biggr\}^{-1}.\end{aligned}\ ] ] the phase space volume @xmath183 which we substitute to find the total phase space flux @xmath184 in the asymptotic limit , this becomes the rate of change of the phase space density , equation ( [ eq : qld ] ) .
the rate of change in the distribution function due to external heating is given by equation ( [ eq : qld ] ) .
we write this in finite difference form for consistency with the fokker - planck scheme and solve after each diffusion step .
the numerical implementation uses a flux - conserving finite - difference scheme with explicit time advance : @xmath185 where the flux @xmath186 is denoted @xmath187 and we have rewritten the previously defined heating rate as @xmath188 the function @xmath189 represents the sum over all resonances which couple to orbits of energy @xmath59 .
in equation ( [ eq : edot ] ) , @xmath190-functions denote resonant coupling of internal and external orbital frequencies . however , the resonant interaction has finite duration due to non - linear saturation or detuning which corresponds to a width in frequency space . for weak perturbations ,
narrow resonances develop since orbital frequencies evolve slowly . for strong perturbations
large widths occur because frequencies evolve rapidly .
the grid spacing employed in the difference scheme defines the frequency widths .
wider spacing implies broader resonances .
broader resonances reduce the heating rate by smearing the input power over a wide region in phase space as shown in figure [ fig : gridding ] . to estimate the proper grid spacing
, we use the bandwidth theorem or uncertainty relation ( e.g bracewell 1986 ) .
the bandwidth theorem identifies the reciprocal relationship between the frequency width and sampling time of an oscillator : @xmath191 here the sampling time is the duration of resonance .
since non - linearity develops with some typical change in energy @xmath192 , we use the rate of energy input to estimate the duration of resonance : @xmath193 comparisons with simulation indicate that 10% change in energy typically leads to frequency evolution .
we calibrate the appropriate frequency widths using fully self - consistent n - body calculations described in appendix [ sec : compsim ] . for typical heating rates in
tidally limited clusters , an initial spacing of @xmath194 is appropriate .
we use this value for all tidally - limited calculations .
larger tidal truncations require an increased width .
two final implementation issues are the boundary conditions on the external heating equation and the fokker - planck equation . for the boundary conditions on equation ( [ eq : a1 ] )
, we set the flux to zero at the center and the gradient of the flux to zero at the edge .
the latter condition represents evaporation .
the last grid point of the df therefore stays fixed between each diffusion step .
we tested the choice in outer boundary condition using a zero flux condition and found solutions which differed by a few percent at most .
we use the standard tidal boundary condition in the fokker - planck equation .
this calls for truncating the distribution function at the maximum energy allowed by the tidal limit .
because strict application of the boundary condition calls for truncating the cluster at @xmath45 , we would throw out a potentially large fraction of the initial cluster distribution in cases where @xmath49 . as a less extreme alternative
, we place the zero - df boundary at the initial limiting energy of the cluster and set the heating rate according to our choice of @xmath45
. then we allow the boundary to evolve according to the mass loss .
tests with n - body calculations shown in appendix [ sec : compsim ] indicate that the prescription works correctly .
comparisons of the external heating theory with n - body simulations are used to test linearity , the assumption of isotropy over long times , the importance of non - spherical moments in the potential , and the boundary condition on the heating equation ( equation [ eq : a1 ] ) and to calibrate frequency widths of heating rates .
we use a self - consistent field expansion code ( e.g. hernquist & ostriker 1992 ) with @xmath195 particles , radial expansion order @xmath196 and angular order @xmath197 .
figures [ fig : k5.c2.comp.nb ] and [ fig : k5.e3.comp.nb ] show comparisons for @xmath84 and @xmath83 orbits .
the two methods agree fairly well , especially at early times , before non - linearity and relaxation produce differences at later times . | we study the combined effects of relaxation , tidal heating and binary heating on globular cluster evolution , exploring the physical consequences of external effects and examining evolutionary trends in the milky way population .
our analysis demonstrates that heating on circular and low - eccentricity orbits can dominate cluster evolution .
the results also predict rapid evolution on eccentric orbits either due to strong relaxation caused by the high densities needed for tidal limitation or due to efficient bulge shocking of low density clusters .
the combination of effects leads to strong evolution of the population as a whole .
for example , within the solar circle , tidally - limited @xmath0 clusters lose at least 40% of their mass in 10 gyr . at high eccentricity
most of these clusters evaporate completely .
bulge shocking disrupts clusters within 40 kpc which have less than 80% of their mass within their pericentric inner lagrange point .
our results are consistent with suggestions that the shape of the cluster luminosity function results from evaporation and disruption of low mass clusters ; they further predict that the net velocity dispersion of the cluster system in the inner galaxy has decreased with time .
preliminary constraints on formation models are also discussed .
we conclude that the observed cluster system has largely been shaped by dynamical selection .
= -0.7 in # 1#2#1 # 2 # 1#2d # 1d # 2 2#1#2d^2 # 1d#2 ^ 2 # 1to 0pt#1 globular clusters : general the galaxy galaxies : star clusters
stellar dynamics |
the classification of the smooth projective varieties satisfying at least one laplace equation is a classical problem , still very far from being solved .
we recall that a projective variety @xmath8 is said to satisfy a laplace equation of order @xmath0 , for an integer @xmath9 , if its @xmath0-osculating space at a general point has dimension strictly less than expected .
the most famous example is the togliatti surface , a rational surface in @xmath10 parametrized by cubics , obtained from the @xmath11rd veronese embedding @xmath12 of @xmath13 by a suitable projection from four points : the del pezzo surface obtained projecting @xmath12 from three general points on it admits a point which belongs to all its osculating spaces , so projecting further from this special point one obtains a surface having all osculating spaces of dimension @xmath14 instead of the expected @xmath15 .
this surface is named from eugenio togliatti who gave a classification of rational surfaces parametrized by cubics and satisfying at least one laplace equation of order @xmath16 . for more details
see the original articles of togliatti @xcite , @xcite , or @xcite , @xcite , @xcite for discussions of this example . in @xcite
the two authors of this note and ottaviani described a connection , due to apolarity , between projective varieties satisfying at least one laplace equation and homogeneous artinian ideals in a polynomial ring , generated by polynomials of the same degree and failing the weak lefschetz property ( wlp for short ) .
let us recall that a homogeneous ideal @xmath17 $ ] fails the weak lefschetz property in some degree @xmath18 if , for any linear form @xmath19 , the map of multiplication by @xmath19 from @xmath20 to @xmath21 is not of maximal rank ( see @xcite ) .
thanks to this connection , explained in detail in section 2 , they obtained in the toric case the classification of the smooth rational threefolds parametrized by cubics and satisfying a laplace equation of order 2 , and gave a conjecture to extend it to varieties of any dimension .
this conjecture has been recently proved in @xcite .
note that the assumption that the variety is toric translates in the fact that the related ideals are generated by monomials , which simplifies apolarity and allows to exploit combinatorial methods .
this point of view had been introduced by perkinson in @xcite , and applied to the classification of toric surfaces and threefolds satisfying laplace equations under some rather strong additional assumptions on the osculating spaces . in this note
we begin the study of the analogous problems for smooth toric rational varieties parametrized by monomials of degree @xmath22 , or equivalently for artinian ideals of @xmath23 generated by monomials of degree @xmath0 .
the picture becomes soon much more involved than in the case of cubics , and for the moment a complete classification appears out of reach .
we consider mainly minimal smooth toric togliatti systems of forms of degree @xmath0 in @xmath23 , i.e. homogeneous artinian ideals generated by monomials failing the wlp , minimal with respect to this property , and such that the apolar linear system parametrizes a smooth variety .
the first goal of this note is to establish minimal and maximal bounds , depending on @xmath24 and @xmath2 , for the number of generators of togliatti systems of this form , and to classify the systems reaching the minimal bound , or close to reach it .
we then investigate if all values comprised between the minimal and the maximal bound can be obtained as number of generators of a minimal smooth togliatti system .
we prove that the answer is positive if @xmath4 , but negative if @xmath6 .
if we avoid smoothness assumption , the answer becomes positive for @xmath25 but is still negative for @xmath26 , even we detect some intervals and sporadic values that are reached .
finally , as applications of our results , we study the mumford
takemoto stability of the syzygy bundle associated to a minimal smooth togliatti system with @xmath4 .
next we outline the structure of this note . in section [ defs and prelim results ]
we fix the notation and we collect the basic results on laplace equations and the weak lefschetz property needed in the sequel .
section [ minimalnumbergenerators ] contains the main results of this note .
precisely , after recalling the results for degree @xmath16 and @xmath11 , in theorem [ mainthm1 ] we prove that the minimal bound @xmath27 on the number of generators of a minimal smooth togliatti system of forms of degree @xmath0 in @xmath1 variables , for @xmath28 , is equal to @xmath29 , and classify the systems reaching the bound .
then in theorem [ mainthm2 ] we get the complete classification for systems with number of generators @xmath30 .
we also compute the maximal bound @xmath31 and give various examples . in section [ number ]
we prove that for @xmath4 and any @xmath28 all numbers in the range between @xmath27 and @xmath31 are reached ( proposition [ interval ] ) , while for @xmath6 the value @xmath32 is a gap ( proposition [ 2n+3 ] ) .
we then prove that , avoiding smoothness , for @xmath25 the whole interval is covered .
finally section [ associatedbundles ] contains the results about stability of the syzygy bundle for minimal smooth monomial togliatti systems in @xmath11 variables
. * notation . * throughout this work @xmath33 will be an algebraically closed field of characteristic zero and @xmath34)$ ] .
we denote by @xmath35 the veronese variety image of the projective space @xmath36 via the @xmath0-tuple veronese embedding .
@xmath37 stands for the ideal generated by @xmath38 , while @xmath39 denotes the @xmath33-vector space they generate
. * acknowledgement . * part of this work was done while the second author was a guest of the university of trieste and she thanks the university of trieste for its hospitality .
the authors wish to thank the referee for some useful remarks .
in this section , we recall some standard terminology and notation from commutative algebra and algebraic geometry , as well as some results needed later on .
in particular , we briefly recall the relationship between the existence of homogeneous artinian ideals @xmath40 $ ] which fail the weak lefschetz property ; and the existence of ( smooth ) projective varieties @xmath8 satisfying at least one laplace equation of order @xmath41 . for more details , see @xcite and @xcite .
* a. the weak lefschetz property .
* let @xmath42=\oplus _
tr_t$ ] be the graded polynomial ring in @xmath1 variables over the field @xmath33 .
[ def of wlp]let @xmath43 be a homogeneous artinian ideal .
we say that @xmath44 has the _ weak lefschetz property _
( wlp , for short ) if there is a linear form @xmath45 such that , for all integers @xmath18 , the multiplication map @xmath46 has maximal rank , i.e. it is injective or surjective .
we will often abuse notation and say that the ideal @xmath47 has the wlp . in this case
, the linear form @xmath19 is called a _ lefschetz element _ of @xmath44 .
if for the general form @xmath45 and for an integer number @xmath18 the map @xmath48 has not maximal rank we will say that the ideal @xmath47 fails the wlp in degree @xmath18 . the lefschetz elements of @xmath44 form a zariski open , possibly empty , subset of @xmath49 .
part of the great interest in the wlp stems from the ubiquity of its presence ( see , e.g. , @xcite , @xcite , @xcite , @xcite , @xcite - @xcite ) and the fact that its presence puts severe constraints on the possible hilbert functions , which can appear in various disguises ( see , e.g. , @xcite ) .
though many algebras are expected to have the wlp , establishing this property is often rather difficult . for example , it was shown by r. stanley @xcite and j. watanabe @xcite that a monomial artinian complete intersection ideal @xmath43 has the wlp . by semicontinuity , it follows that a _ general _
artinian complete intersection ideal @xmath50 has the wlp but it is open whether _ every _ artinian complete intersection of height @xmath51 over a field of characteristic zero has the wlp .
it is worthwhile to point out that the weak lefschetz property of an artinian ideal @xmath47 strongly depends on the characteristic of the ground field @xmath33 and , in positive characteristic , there are examples of artinian complete intersection ideals @xmath52 $ ] failing the wlp ( see , e.g. , remark 7.10 in @xcite ) .
in @xcite , mezzetti , mir - roig , and ottaviani showed that the failure of the wlp can be used to construct ( smooth ) varieties satisfying at least one laplace equation of order @xmath41 ( see also @xcite and @xcite ) .
let us review the needed concepts from differential geometry in order to state this result .
* b. laplace equations .
* let @xmath8 be a projective variety of dimension @xmath24 and let @xmath53 be a smooth point .
we choose a system of affine coordinates and an analytic local parametrization @xmath54 around @xmath55 where @xmath56 and the @xmath57 components of @xmath58 are formal power series .
the _ @xmath59-th osculating space _
@xmath60 to @xmath61 at @xmath55 is the projectivised span of all partial derivatives of @xmath58 of order @xmath62 .
the expected dimension of @xmath60 is @xmath63 , but in general @xmath64 ; if strict inequality holds for all smooth points of @xmath61 , and @xmath65 for a general point @xmath55 , then @xmath61 is said to satisfy @xmath66 laplace equations of order @xmath59 .
[ bound ] it is clear that if @xmath67 then @xmath61 satisfies at least one laplace equation of order @xmath59 , but this case is not interesting and will not be considered in the following .
let @xmath47 be an artinian ideal generated by @xmath68 homogeneous polynomials @xmath69 of degree @xmath0 .
associated to @xmath70 there is a morphism @xmath71 note that @xmath72 is everywhere regular because @xmath47 is an artinian ideal .
its image @xmath73 is the projection of the @xmath24-dimensional veronese variety @xmath35 from the linear system @xmath74 where @xmath75 is the ideal generated by the macaulay inverse system of @xmath47 ( see @xcite , 3 for details ) .
analogously , associated to @xmath76 there is a rational map @xmath77 the closure of its image @xmath78 is the projection of the @xmath24-dimensional veronese variety @xmath35 from the linear system @xmath79 . the varieties @xmath80 and @xmath81 are usually called apolar . in the following @xmath81
will simply be denoted by @xmath61 .
we have : [ teathm ] let @xmath43 be an artinian ideal generated by @xmath68 homogeneous polynomials @xmath82 of degree @xmath0 .
if @xmath83 , then the following conditions are equivalent : * the ideal @xmath47 fails the wlp in degree @xmath84 ; * the homogeneous forms @xmath82 become @xmath33-linearly dependent on a general hyperplane @xmath85 of @xmath36 ; * the @xmath24-dimensional variety @xmath86 satisfies at least one laplace equation of order @xmath84
. see ( * ? ? ?
* theorem 3.2 ) . in view of remark
[ bound ] , the assumption @xmath87 ensures that the laplace equations obtained in ( 3 ) are not obvious . in the particular case
@xmath4 , this assumption gives @xmath88 .
the above result motivates the following definition : [ togliattisystem ] let @xmath43 be an artinian ideal generated by @xmath68 forms @xmath82 of degree @xmath0 , @xmath83 .
we introduce the following definitions : * @xmath47 is a _ togliatti system _ if it satisfies the three equivalent conditions in theorem [ teathm ] .
* @xmath47 is a _
monomial togliatti system _ if , in addition , @xmath47 ( and hence @xmath75 ) can be generated by monomials .
* @xmath47 is a _
smooth togliatti system _ if , in addition , the @xmath24-dimensional variety @xmath61 is smooth . * a monomial togliatti system @xmath47 is said to be _ minimal _ if @xmath47 is generated by monomials @xmath89 and there is no proper subset @xmath90 defining a monomial togliatti system .
the names are in honor of eugenio togliatti who proved that for @xmath4 the only smooth togliatti system of cubics is @xmath91 $ ] ( see @xcite , @xcite , @xcite ) .
the main goal of our note is to determine a lower bound @xmath92 ( resp .
@xmath93 ) for the minimal number of generators @xmath94 of any ( resp . smooth )
minimal monomial togliatti system @xmath95 $ ] of forms of degree @xmath2 and classify _ all _ ( resp .
smooth ) minimal monomial togliatti systems @xmath95 $ ] of forms of degree @xmath2 which reach the bound , i.e. @xmath96 ( resp .
@xmath97 ) . these results will be achieved in the next section .
from now on , we restrict our attention to monomial artinian ideals @xmath98 $ ] ( i.e. the ideals invariants for the natural toric action of @xmath99 ) . recall that when @xmath43 is an artinian monomial ideal , the homogeneous part @xmath100 of degree @xmath0 of the inverse system @xmath75 is spanned by the monomials in @xmath101 not in @xmath47 .
it is also worthwhile to recall that for monomial artinian ideals to test the wlp there is no need to consider a general linear form .
in fact , we have [ lem - l - element ] let @xmath102 $ ] be an artinian monomial ideal .
then @xmath44 has the wlp if and only if @xmath103 is a lefschetz element for @xmath44 .
see @xcite ; proposition 2.2 .
given an artinian ideal @xmath104 $ ] , we denote by @xmath105 the minimal number of generators of @xmath47 .
we define @xmath106 @xmath107 @xmath108 @xmath109 where @xmath110 is the set of all minimal monomial togliatti systems @xmath104 $ ] of forms of degree @xmath0 and @xmath111 is the set of all minimal smooth monomial togliatti systems @xmath104 $ ] of forms of degree @xmath0 . by definition , we have @xmath112 . our first goal is to provide a lower bound for @xmath113 and @xmath93 .
first , we observe that all artinian monomial ideals @xmath95 $ ] generated by forms of degree @xmath2 contain @xmath114 for @xmath115 and the ideals @xmath116 do satisfy wlp .
therefore , we always have @xmath117 let us start analyzing the cases @xmath118 .
the minimal smooth monomial togliatti systems @xmath95 $ ] of quadrics were classified in @xcite ; proposition 2.8 .
it holds : * @xmath119 . * for @xmath120 , we have @xmath121 * for @xmath120 , @xmath122 .
in particular , for @xmath25 we have @xmath123 ; for @xmath124 we have @xmath125 ; and for all @xmath126 the inequalities in ( [ ineq ] ) are strict , i.e. , @xmath127 we also have @xmath128 for @xmath7 ( since we easily check that @xmath129 and @xmath130 fails weak lefschetz property from degree 1 to degree 2 ) and @xmath131 ( since @xmath132 and @xmath133 fails weak lefschetz property in degree 1 ) .
the minimal smooth monomial togliatti systems @xmath95 $ ] of cubics were classified in @xcite ; theorem 4.11 and @xcite ; theorem 3.4 .
it holds : * @xmath134 , * @xmath135 , * @xmath136 , and * for all @xmath7 , we have @xmath137,@xmath138 @xmath139 and , hence @xmath140 we may also check that @xmath141 for @xmath120 ( since @xmath142 and @xmath143 fails weak lefschetz property in degree 2 ) and @xmath144 ( since @xmath145 and @xmath146 fails weak lefschetz property from degree 2 to degree 3 ) .
notice that @xmath147 unless @xmath148 and @xmath149 unless @xmath5 .
from now on , we assume @xmath150 and @xmath151 .
we will prove that @xmath152 .
in addition , we will classify all ( resp .
smooth ) minimal monomial togliatti systems @xmath153 $ ] of forms of degree @xmath150 with @xmath154 and all smooth minimal monomial togliatti systems @xmath104 $ ] of forms of degree @xmath150 with @xmath155 , revealing how the power of combinatorics tools can allow us to deduce pure geometric properties of projections of @xmath24-dimensional veronese varieties @xmath35 . to prove it
, we will associate to any artinian monomial ideal a polytope and the toric variety @xmath86 introduced in 2 b. hence , we will be able to tackle our problem with tools coming from combinatorics .
in fact , when we deal with artinian monomial ideals @xmath156 $ ] the failure of the wlp can be established by fairly easy combinatoric properties of the associated polytope @xmath157 . to state
this result we need to fix some extra notation .
let @xmath156 $ ] be an artinian monomial ideal generated by monomials of degree @xmath0 and let @xmath75 be its inverse system .
we denote by @xmath158 the standard @xmath24-dimensional simplex in the lattice @xmath159 , we consider @xmath160 and we define the polytope @xmath157 as the convex hull of the finite subset @xmath161 corresponding to monomials of degree @xmath0 in @xmath75 . as usual
we define the sublattice @xmath162 in @xmath159 generated by @xmath163 as follows : @xmath164 we have : [ failurewlp ] let @xmath156 $ ] be an artinian monomial ideal generated by @xmath68 monomials of degree @xmath0 .
assume @xmath83 .
then , @xmath47 is a togliatti system if and only if there exists a hypersurface of degree @xmath84 containing @xmath165 .
in addition , @xmath47 is a minimal togliatti system if and only if any such hypersurface @xmath166 does not contain any integral point of @xmath167 except possibly some of the vertices of @xmath168 .
it follows from theorem [ teathm ] and @xcite , proposition 1.1 .
let us illustrate the above proposition with a precise example .
the artinian ideal @xmath169 $ ] defines a minimal monomial togliatti system of cubics .
in fact , the set @xmath170 is : @xmath171 @xmath172 there is a hyperquadric , and only one , containing all points of @xmath163 and no integral point of @xmath173 , namely , @xmath174 for seek of completeness we also recall the following useful combinatorial criterion which will allow us to check if a subset @xmath175 of points in the lattice @xmath159 defines a smooth toric variety @xmath176 or not . [ smoothness ] let @xmath156 $ ] be an artinian monomial ideal generated by monomials of degree @xmath0 .
let @xmath165 be the set of integral points corresponding to monomials in @xmath76 , @xmath177 the semigroup generated by @xmath163 and 0 , @xmath157 the convex hull of @xmath163 and @xmath178 the projective toric variety associated to the polytope @xmath157 .
@xmath178 is smooth if and only if for any non - empty face @xmath179 of @xmath157 the following conditions hold : \(i ) the semigroup @xmath180 is isomorphic to @xmath181 with @xmath182 .
\(ii ) the lattices @xmath183 and @xmath184 coincide .
see @xcite ; chapter 5 , corollary 3.2 .
note that in this case @xmath185 .
figure 1 illustrates two examples of minimal togliatti systems which are non - smooth .
the points of the complementary of @xmath163 are marked with a cross .
[ 1 ] and @xmath186,title="fig:",width=151 ] and @xmath186,title="fig:",width=151 ] the condition ( i ) of proposition [ smoothness ] is verified if and only if translating each vertex @xmath187 of the polygon to the origin of @xmath188 and considering for each edge coming out of @xmath187 the first point with integer coordinates , these form a @xmath189-basis of @xmath188 .
the condition ( ii ) is equivalent to require that each point of @xmath188 which lies on an edge of the polygon is also a point of @xmath163 .
therefore , the first figure violates condition ( i ) and the second one violates condition ( ii ) . in order to achieve the classification of minimal ( resp .
smooth ) monomial togliatti systems @xmath190 $ ] of degree @xmath150 with @xmath94 as small as possible we need to introduce one more definition .
a togliatti system @xmath156 $ ] of forms of degree @xmath0 is said to be _ trivial _ if there exists a form @xmath166 of degree @xmath84 such that @xmath47 contains @xmath191 .
the following remark justifies why we call them trivial .
\(i ) let @xmath166 be a homogeneous form of degree @xmath84 .
since @xmath192 become linearly dependent on the hyperplane @xmath193 , using proposition [ lem - l - element ] , we conclude that any artinian ideal of the form @xmath194 is a ( trivial ) togliatti system . in the monomial case , looking at the inverse system that parameterizes the surface @xmath61
, we can observe that it satisfies a laplace equation of the simplest form , given by the annihilation of the partial derivative of order @xmath84 corresponding to the monomial @xmath166 .
\(ii ) let @xmath104 $ ] be a monomial togliatti system of cubics .
if @xmath47 is trivial then it is not smooth .
[ mainthm1 ] for any integer @xmath151 and @xmath150 , we have @xmath195 . in particular , if @xmath196 $ ] is a minimal ( resp .
smooth minimal ) monomial togliatti system of forms of degree @xmath0 , then @xmath197 .
in addition , _ all _ minimal monomial togliatti systems @xmath198 $ ] of forms of degree @xmath150 with @xmath154 are trivial unless one of the following cases holds : * @xmath199 and , up to a permutation of the coordinates , @xmath200 . * @xmath201 and , up to a permutation of the coordinates , @xmath202 .
furthermore , ( i ) is smooth and ( ii ) is not smooth .
first of all we observe that @xmath203 $ ] is a minimal monomial togliatti system of forms of degree @xmath0 and by proposition [ smoothness ] , being @xmath28 , it is smooth .
thus , @xmath204 . to prove that @xmath205 , we have to check that any monomial artinian ideal @xmath206 with @xmath207 , @xmath208 has the wlp at the degree @xmath84 . according to proposition [ failurewlp ] , to prove the last assertion it is enough to prove that no hypersurface of degree @xmath84 contains all points of @xmath165 , where as before @xmath161 is the set of all integral points corresponding to monomials of degree @xmath0 in @xmath75 .
for any integer @xmath209 , we set @xmath210 and @xmath211 ; we have @xmath212 . to illustrate this method , in figure 2 we show the pictures of the sets @xmath163 , and @xmath213 , @xmath214 , @xmath215 , @xmath216 , when @xmath217 .
[ 2 ] with @xmath218,title="fig:",width=151 ] @xmath213 @xmath214 @xmath215 @xmath216 we will prove now the theorem proceeding by induction on @xmath24 .
let us start with the case @xmath4 .
we take a monomial artinian ideal @xmath219 with @xmath220 and we show that no plane curve of degree @xmath84 contains all points of @xmath221 . since @xmath222 , we can assume wlog that @xmath223 .
we assume that there is a plane curve @xmath224 of degree @xmath84 containing all points of @xmath163 and we will get a contradiction . since @xmath224 contains the @xmath0 points of @xmath214 , it factorizes as @xmath225 . since @xmath226 contains the @xmath84 points @xmath213 , it factorizes as @xmath227 . now , if @xmath228 , then @xmath229 contains @xmath230 points , if @xmath231 , then @xmath229 contains @xmath84 points ; in any case @xmath232 for a suitable form @xmath233 of degree @xmath234 . repeating the argument
we get that @xmath235 , so @xmath224 does not contain the points of @xmath236 , which is non - empty by assumption .
this contradicts the existence of a plane curve of degree @xmath84 containing all integral points of @xmath163 .
let now @xmath6 and assume that the claim is true for @xmath237 .
let us prove that no hypersurface of degree @xmath84 contains all points of @xmath165 , where @xmath238 with @xmath207 , @xmath208 .
wlog we can assume @xmath239 and also @xmath240 .
therefore @xmath241 , so @xmath242 appears explicitly in the monomial @xmath243 and @xmath244 is equal to @xmath245 minus the @xmath24 vertices and at most @xmath246 other points . by inductive assumption ,
no hypersurface in @xmath24 variables of degree @xmath84 contains @xmath244 , so @xmath224 factorizes as @xmath247 , where @xmath226 is a hypersurface of degree @xmath230 containing all points of @xmath248 .
if the @xmath237 monomials have @xmath249 , then @xmath250 and we deduce that @xmath251 , because for @xmath252 the simplex @xmath253 is not contained in any hypersurface in @xmath237 variables of degree @xmath254 .
this gives a contradiction because @xmath224 misses all points of @xmath255 .
otherwise , @xmath256 minus at most @xmath246 points .
then by inductive assumption there is no hypersurface of degree @xmath84 in @xmath237 variables containing @xmath257 .
then we repeat the argument until we reach a contradiction .
finally we will classify all minimal monomial togliatti systems @xmath190 $ ] of forms of degree @xmath150 with @xmath154 .
first we assume that @xmath4 and we will show that all of them are trivial unless @xmath186 and @xmath200 or @xmath258 and @xmath202 .
take @xmath259 $ ] with @xmath260 and @xmath261 a minimal togliatti system .
if there exists @xmath262 such that @xmath263 ( wlog we assume @xmath264 ) then the plane curve @xmath224 containing all integral points of @xmath163 factorizes @xmath265 and since @xmath224 can not miss any point of @xmath163 we must have @xmath266 which forces @xmath267 , @xmath268 .
assume now that for any @xmath269 , there exists @xmath270 with @xmath271 .
since @xmath150 we may assume @xmath272 and @xmath273 .
therefore , @xmath274 and @xmath275 .
but none gives a minimal togliatti system because @xmath276 are linearly independent on a general line of @xmath13 ( see theorem [ teathm ] ) unless @xmath186 and @xmath277 and @xmath278 or @xmath258 and @xmath279 and @xmath280 .
furthermore , applying proposition [ smoothness ] , we easily check that only @xmath200 defines a smooth variety .
assume now @xmath120 and @xmath281 and let @xmath282 $ ] with @xmath283 and @xmath284 be a togliatti system .
there is an integer @xmath18 , @xmath285 such that @xmath286 .
therefore , wlog we can assume @xmath287 .
arguing as in the previous part of the proof any hypersurface @xmath224 of degree @xmath84 containing all integral points of @xmath163 factorizes @xmath265 and since @xmath224 can not miss any point of @xmath163 we must have @xmath266 which forces @xmath267 , @xmath288 and hence @xmath47 is trivial , which proves what we want .
minimal monomial togliatti systems @xmath289 $ ] of forms of degree @xmath150 with @xmath290 were also classified by albini in @xcite ; theorem 3.5.1 .
so , our results can be seen as a generalization of his result to the case of an arbitrary number of variables .
[ triv ] up to permutation of the variables , the trivial togliatti systems with @xmath154 are of the form @xmath291 . figure [ 3 ] illustrates the only smooth non
trivial example of minimal togliatti system of forms of degree @xmath15 with @xmath290 .
[ 3 ] and @xmath186,title="fig:",width=151 ] in the case of the non - trivial minimal smooth monomial togliatti system @xmath292 all @xmath293-osculating spaces to @xmath61 have dimension lower than @xmath294 , which is the expected dimension , but the dimension of the previous osculating spaces is not constant . some points of @xmath61 have @xmath16-osculating space or @xmath11-osculating space of dimension less than the general one ( they are _ flexes _ of @xmath61 ) .
this follows from @xcite , where it is proved that the dimension of the @xmath59-osculating space at a point @xmath53 , corresponding to a vertex @xmath295 of the polytope @xmath157 , is maximal if and only if @xmath296 contains all points out to level @xmath297 with respect to @xmath295 .
this means that , after translating @xmath295 to the origin and using the first lattice points lying along the two edges of @xmath157 emanating from @xmath295 as basis for the lattice , @xmath298 contains all points @xmath299 with @xmath300 .
this remark explains why this example is not included in the list of perkinson @xcite , theorem 3.2 . to better understand its geometry ,
let us note that the surface @xmath61 is the projection , from a line @xmath19 , of the blowing up of @xmath13 at three general points @xmath301 , @xmath302 , @xmath303 , embedded in @xmath304 by the linear system of the quintics through them .
the line @xmath19 is chosen so to meet all @xmath293-osculating spaces of this surface .
we observe that there are three lines of this type , obtained by interchanging the variables .
every such line meets also the @xmath11-osculating space at one of the three points @xmath305 , and the @xmath16-osculating spaces at the other two .
this gives rise to the flexes .
any curve on @xmath61 corresponding to a general line through one of the blown up points is a smooth rational quartic .
one can check that the flexes result to be singular points of intersection of two irreducible components of some reducible quartics obtained after the projection from @xmath19 .
it would be nice to have a precise geometric description of the inflectional loci of @xmath61 , but this goes beyond the scope of this article , we plan to return on this topic in a forthcoming paper . the hypersurface @xmath224 of degree @xmath84 that contains the integral points @xmath163 of a minimal monomial togliatti system @xmath306\ ] ] with @xmath154 can be described .
it turns out that if @xmath47 is trivial then @xmath224 is the union of @xmath84 hyperplanes . if @xmath4 , @xmath258 and @xmath307 , then @xmath308 . in this example
the surface @xmath309 is the closure of the image of the parametrization @xmath310 defined by the monomials of degree @xmath293 not in @xmath47 , i.e. @xmath311 one computes that its partial derivatives of order @xmath11 satisfy the laplace equation @xmath312 if @xmath4 , @xmath186 and @xmath200 then @xmath313 which is irreducible .
similarly , the laplace equation satisfied by the parametrization of the surface @xmath314 is @xmath315 @xmath316 fix integers @xmath150 and @xmath151 .
let @xmath317 $ ] be a monomial artinian ideal of forms of degree @xmath0 . if @xmath318 then , for any @xmath319 , the @xmath59-osculating space to @xmath61 at a general point @xmath53 has the expected dimension , namely @xmath320 . in next theorem
we will classify all smooth minimal monomial togliatti systems @xmath321 whose minimal number of generators exceed by one the possible minimum .
we start with a lemma .
[ l_0 ] let @xmath322 $ ] with @xmath323 , @xmath324 for @xmath325 , be a minimal togliatti system of forms of degree @xmath326 .
assume @xmath327 .
if @xmath328 , then @xmath329 for all index @xmath330 . since @xmath47 is a togliatti system , there exists a form @xmath224 of degree @xmath84 in @xmath331 passing through all points of @xmath163 .
its restriction to @xmath332 , @xmath333 , vanishes at all points of @xmath213 . by assumption , to get @xmath213 we have to remove from the simplex @xmath245 the @xmath24 vertices and at most @xmath246 other points .
we denote by @xmath334 $ ] the ideal generated by @xmath335 and the monomials not containing @xmath242 among @xmath336 .
if @xmath337 , @xmath338 is a togliatti system in @xmath24 variables with @xmath339 , which contradicts theorem [ mainthm1 ] .
hence @xmath340 and @xmath341 . but @xmath47 is minimal , so by proposition [ failurewlp ]
@xmath342 does not contain any point of @xmath167 except the vertices , which implies that @xmath329 for any index @xmath330 .
[ trivialb ] recall that , when @xmath47 is a monomial togliatti system , the projective variety @xmath61 defined by the apolar linear system of forms of degree @xmath0 has all @xmath343-osculating spaces of dimension strictly less than expected , i.e. @xmath61 satisfies a laplace equation of order @xmath84 .
since the @xmath343-osculating spaces of @xmath35 have the expected dimension , this means that the space that @xmath47 determines meets the @xmath343-osculating space @xmath344 for all @xmath345 . as pointed out in @xcite ,
4 , when @xmath47 is as in lemma [ l_0 ] , i.e. all monomials in @xmath47 except @xmath346 are multiple of one variable , there is a point @xmath347 such that the intersection of @xmath47 with the @xmath343-osculating space at @xmath348 meets all the other @xmath343-osculating spaces .
these togliatti systems were called in @xcite trivial of type b. for instance , if @xmath349 and @xmath350 are any general monomials of degree @xmath84 , the ideal @xmath351 is a minimal togliatti system of the type just described .
[ mainthm2 ] let @xmath104 $ ] be a smooth minimal monomial togliatti system of forms of degree @xmath150 .
assume that @xmath352 . then @xmath47 is trivial unless @xmath4 and , up to a permutation of the coordinates , one of the following cases holds : * @xmath186 and @xmath353 or @xmath354 or @xmath355 .
* @xmath356 and @xmath357 or @xmath358 or @xmath359 .
let us first assume that @xmath4 and let @xmath360 $ ] with @xmath260 and @xmath261 be a minimal smooth togliatti system .
we distinguish several cases : we assume that there is @xmath361 such that @xmath362 .
wlog we can assume @xmath363 and @xmath364 .
let @xmath224 be a plane curve containing all points of @xmath163 . since @xmath224 contains the @xmath0 points of @xmath214 and the @xmath84 points of @xmath213 it factorizes as @xmath227 .
let @xmath365 , then @xmath366 contains @xmath367 integral points of @xmath368 ; to get @xmath369 we have to remove three points , the first one from @xmath370 , the second one from @xmath371 , and the third one from @xmath372 .
first of all we want to exclude that @xmath373 .
otherwise @xmath374 has as factors @xmath375 ; but in view of minimality @xmath370 must be contained in @xmath163 , which gives a contradiction . therefore @xmath376 and there are two subcases to analyze separately : * @xmath377 . in this case
@xmath224 factorizes as @xmath378 and the plane curve @xmath224 contains all points of @xmath163 if and only if @xmath379 .
but in this case @xmath380 , @xmath381 , @xmath382 , and applying proposition [ smoothness ] we deduce that @xmath383 is not a smooth togliatti system since it violates condition ( ii ) of proposition [ smoothness ] . * @xmath384 . in this case @xmath227 and @xmath374 contains all integral points in @xmath385 if and only if @xmath386 and @xmath387 for a suitable @xmath388 .
therefore , @xmath47 is a trivial smooth togliatti system .
we assume @xmath389 for all @xmath390 and that for all @xmath391 there exists @xmath392 such that @xmath393 .
we distinguish 4 subcases and a straightforward computation allows us to conclude : * @xmath394 is a smooth minimal togliatti system if and only if @xmath186 or 7 .
* @xmath395 with @xmath396 is a smooth minimal togliatti system if and only if @xmath186 and @xmath397 . *
@xmath398 with @xmath399 and @xmath400 is a smooth minimal togliatti system if and only if @xmath356 , @xmath401 and @xmath402 .
* @xmath403 is a smooth minimal togliatti system if and only if @xmath186 and @xmath404 or @xmath356 and @xmath405 .
we assume that there exists @xmath406 , @xmath389 for all @xmath407 and that for all @xmath391 there exists @xmath392 such that @xmath393 .
the smoothness criterion ( proposition [ smoothness ] ) implies that , up to permutation of the coordinates , we have @xmath408 and we can assume @xmath409 and @xmath410 with @xmath411 and @xmath47 is never a smooth minimal togliatti system .
we assume that there exists @xmath412 , and that for all @xmath391 there exists @xmath413 such that @xmath414 .
the smoothness criterion ( proposition [ smoothness ] ) implies that , up to permutation of the coordinates , we have @xmath408 , @xmath268 and @xmath415 which does not correspond to a smooth minimal togliatti system .
let us now assume that @xmath6 .
we want to prove that all minimal smooth monomial togliatti systems @xmath190 $ ] of forms of degree @xmath150 with @xmath352 are trivial .
this time we distinguish two cases : for all @xmath416 , @xmath417 .
this implies that each variable @xmath418 appears explicitly in exactly two of the monomials @xmath419 .
equivalently , looking at the simplex , the @xmath1 integral points to remove from @xmath168 to get @xmath163 are all on the exterior facets , and on each facet there are exactly @xmath237 points .
we consider now the restriction of the hypersurface @xmath224 to a facet , we apply theorem [ mainthm1 ] and we get that the corresponding @xmath237 monomials , together with the @xmath0th powers of the corresponding variables , form a trivial togliatti system in @xmath24 variables , of the form described in remark [ triv ] .
this gives a contradiction , so this case is impossible
. there exists @xmath416 such that @xmath420 .
wlog we can assume @xmath421 and @xmath422 .
therefore , in view of lemma [ l_0 ] @xmath423 .
this means that all monomials @xmath419 contain @xmath242 .
we consider the restrictions of @xmath424 to the hyperplane @xmath425 , they are linearly dependent by assumption .
but in @xmath426 there is some monomial not containing @xmath242 , that can not cancel with the others , so its coefficient in a null linear combination must be @xmath427 , and by consequence also the coefficients of @xmath428 are @xmath427 .
this implies that the monomials @xmath419 divided by @xmath242 , together with @xmath429 form again a togliatti system but of degree one less , with the same properties .
so we can proceed by induction on the degree , until we arrive to @xmath258 .
now we have to prove that there is no hypersurface @xmath430 of degree @xmath11 containing all points of @xmath163 unless @xmath47 is a trivial monomial togliatti system . since @xmath423 , @xmath431 and @xmath432 contains all points of @xmath433 .
this is possible if and only if @xmath47 is trivial of type @xmath434 where @xmath435 is a monomial of degree @xmath84 involving at least @xmath16 variables . in theorem
[ mainthm2 ] , we did not use the smoothness assumption in the cases with @xmath6 . to complete the results of theorems [ mainthm1 ] and [ mainthm2 ] , in next proposition
we give a criterion to distinguish the smooth ones among the trivial togliatti systems .
to have a complete picture we also include systems with number of generators bigger than @xmath436 .
[ triv smooth ] let @xmath47 be a trivial togliatti system of the form @xmath437 , where @xmath435 is a monomial .
then @xmath47 is smooth if and only if one of the following happens ( up to permutation of the variables ) : * @xmath438 and @xmath4 or @xmath25 ; * @xmath439 , @xmath4 , @xmath440 ; * @xmath28 , @xmath4 , @xmath441 or @xmath442 with @xmath443 ; * @xmath28 , @xmath6 , @xmath441 or @xmath444 with @xmath445 and @xmath446 .
if @xmath438 , we may assume that @xmath447 . if @xmath4 then @xmath61 is a point .
hence , the system @xmath47 is smooth .
assume @xmath6 .
after cutting the points of @xmath47 from @xmath448 it remains @xmath449 , which is the @xmath450-dimensional simplex minus the @xmath24 vertices . through each vertex of the polytope @xmath157
there are @xmath451 edges . then the system is singular unless @xmath25 .
indeed by proposition [ smoothness ] , ( i ) , for @xmath61 to be smooth the number of edges emanating from each vertex must be equal to @xmath237 . if @xmath439 then @xmath435 can be @xmath452 , or @xmath453 . if @xmath4 , the first case is smooth , because @xmath157 is a trapezium , and the second one is singular : indeed , we cut from @xmath448 the whole edge @xmath454 .
so an edge of @xmath157 is @xmath455 , but the central point @xmath456 does not belong to @xmath163 .
therefore this edge gives a singularity .
if @xmath6 both cases are singular : the first one because through the vertices of @xmath157 adjacent to @xmath457 there are more than @xmath24 edges , the second one because @xmath157 contains the @xmath458-dimensional faces for @xmath459 now assume @xmath28 and @xmath4 .
if @xmath441 , then the system is clearly smooth . if @xmath460 then it is singular because the situation is as in figure 1 . if @xmath461 with @xmath462 , the system is singular because in the edge @xmath463 of @xmath157 we have to cut two points in the middle
finally if @xmath442 , with @xmath464 all strictly positive , we get a smooth system because the points of @xmath47 are all inner points in @xmath465 if @xmath28 and @xmath6 , then if @xmath435 is @xmath466 , the system is smooth ; if @xmath467 or @xmath461 with @xmath462 the system is singular , because @xmath157 has a @xmath16-dimensional face which is singular .
finally if @xmath435 contains at least @xmath11 of the variables the system is smooth : indeed on the @xmath458-dimensional edges of @xmath157 there are no points of @xmath47 , while on the faces of @xmath468 of dimension at least @xmath16 the points of @xmath47 are in the interior .
we consider now the range comprised between @xmath27 and @xmath31 ( resp . @xmath113 and @xmath436 ) and ask if all values are reached .
next proposition gives a rather precise picture in the case @xmath4 .
[ interval ] with notation as in section [ minimalnumbergenerators ] we have : * for any @xmath150 , @xmath469 . * for any @xmath150 , @xmath470 . * for any @xmath150 and any @xmath471 , there exists @xmath472 with @xmath473 .
\(i ) it follows from theorem [ mainthm1 ] .
\(ii ) by definition we have @xmath474 for any @xmath150 .
the inequality @xmath475 ( and , hence , @xmath476 ) will follows from ( iii ) .
\(iii ) for any @xmath150 and for any @xmath477 , we consider the ideals @xmath478 @xmath479 we have @xmath480 and it follows from propositions [ failurewlp ] and [ smoothness ] that @xmath481 $ ] , which proves what we want .
proposition [ interval ] does not generalize to the case @xmath120 , i.e. not all values of @xmath482 , @xmath483 , occur as the minimal number of generators of a smooth togliatti system @xmath484 . the first case is illustrated in next lemma for the case @xmath439 and next proposition for the general case @xmath150 . [ d=3,2n+3 ]
assume @xmath7 and let @xmath47 be a minimal togliatti system of cubics .
then , @xmath485 .
in addition , we have : * @xmath154 if and only if @xmath47 is trivial , i.e. , up to permutations of the coordinates , @xmath486 . in particular , @xmath487 .
* @xmath352 if and only if @xmath47 is trivial , i.e. , up to permutations of the coordinates , @xmath488 with @xmath489 .
in particular , @xmath490 .
* @xmath491 .
we proceed by induction on @xmath24 . with macaulay2 ( @xcite )
we easily check that @xmath492 for any @xmath493 .
assume now @xmath494 and suppose that the result is true for @xmath237 .
we take @xmath495 with @xmath496 , @xmath497 , and we will see that there is no hyperquadrics @xmath432 containing all points of @xmath163 . assume it exists and we will get a contradiction .
wlog we can assume that @xmath242 appears explicitly in the monomial @xmath498 and @xmath213 is equal to @xmath499 minus @xmath24 vertices and at most @xmath246 other points . by induction
no hyperquadric in @xmath500 contains @xmath213 .
so @xmath432 decomposes as @xmath501 and since there is no hyperplane @xmath502 containing all the points of @xmath433 we get a contradiction .
let us now classify all togliatti systems @xmath503 , @xmath7 , with @xmath504 .
\(i ) assume @xmath124 , @xmath493 and @xmath505 . using macaulay2 we get that @xmath47 is trivial .
suppose now @xmath494 , let @xmath506 with @xmath283 and @xmath507 , and let @xmath432 a hyperquadric passing through the points of @xmath163 .
wlog we can assume @xmath287 . therefore , @xmath432 factorizes as @xmath508 and since @xmath509 can not miss any point of @xmath163 we must have @xmath510 which forces @xmath511 and hence @xmath47 is trivial .
\(ii ) using macaulay2 we prove that if @xmath124 , @xmath493 and @xmath512 then @xmath47 is trivial .
suppose now @xmath494 and let @xmath513 with @xmath283 and @xmath507 .
wlog we can assume @xmath514 and @xmath241 .
if @xmath515 then @xmath423 by lemma [ l_0 ] and @xmath501 where @xmath502 is a hyperplane containing all points of @xmath433 .
this is possible if and only if @xmath47 is trivial of type @xmath516 with @xmath489 . if @xmath517 then using hypothesis of induction together with the fact that @xmath241 we get that the restriction of @xmath518 to the hyperplane @xmath519 is trivial of type @xmath520 or @xmath521 with @xmath522 . therefore , either @xmath523 or @xmath524 with @xmath525 ; and none of them belongs to @xmath526 .
\(iii ) again using macaulay 2 we prove that the result is true for @xmath124 .
suppose now @xmath494 and let @xmath527 with @xmath528 and @xmath529 .
wlog we can assume @xmath530 and @xmath241 . if @xmath531 then @xmath423 by lemma [ l_0 ] and @xmath501 but this is impossible since there is no a hyperplane containing all points of @xmath532 and no point of @xmath533 a part from the vertices .
if @xmath534 then using hypothesis of induction together with the fact that @xmath241 we get that the restriction of @xmath535 to the hyperplane @xmath519 is trivial of type @xmath520 or @xmath521 , @xmath522 , or @xmath536 , @xmath537 or @xmath538 , @xmath522 , @xmath539 .
therefore , @xmath540 or @xmath524 , @xmath522 or @xmath541 , @xmath537 or @xmath542 , @xmath522 , @xmath539 ; and none of them belongs to @xmath526 .
[ 2n+3 ] let @xmath6 and @xmath28 .
then there is no @xmath543 with @xmath544 .
we distinguish two cases : 1 . for all @xmath416 , @xmath545
, i.e. every variable appears in at most three of the monomials @xmath546 .
+ if one of the monomials contains all the variables , the other @xmath1 monomials contain two variables each , and we are in the same situation of theorem [ mainthm2 ] , case 1 , which is impossible .
therefore no monomial contains all variables and at least two variables appear in three monomials .
assume that @xmath242 appears in three monomials ; then @xmath224 passes through the integral points of @xmath213 .
recall that @xmath213 is equal to @xmath245 minus the @xmath24 vertices and @xmath237 other points .
so the removed points form a togliatti system @xmath338 in the @xmath24 variables @xmath547 with @xmath548 and we can apply theorem [ mainthm1 ] .
there are two possibilities : * @xmath25 and @xmath338 is one of the two special togliatti systems of degree @xmath15 or @xmath293 of theorem [ mainthm1 ] . if @xmath186 , up to permutation of the variables the only possibility is @xmath549 with @xmath550 .
but it is easy to check that this is not a togliatti system . in the case @xmath258
there are two possibilities : @xmath551 with @xmath550 , @xmath552 , or @xmath553 with @xmath554 .
both systems are not togliatti .
* @xmath338 is of the form @xmath555 . in this case
@xmath556 appears in at least @xmath237 monomials , therefore @xmath25 or @xmath124 .
+ if @xmath25 , the other three monomials in @xmath47 are either of the form @xmath557 , or of the form @xmath558 , with @xmath559 , @xmath560 , @xmath552 .
it is immediate to check that they are not togliatti systems . if @xmath124 then the six monomials @xmath561 are of the form @xmath562 . also in this case the system is not togliatti .
there exists an index @xmath18 such that @xmath563 , i.e. one of the variables appears in at least @xmath293 monomials .
we can assume @xmath363 .
therefore , by lemma [ l_0 ] , @xmath242 appears in all monomials @xmath564 .
let @xmath565 , @xmath566 .
as in the proof of theorem [ mainthm2 ] , case 2 , we observe that @xmath567 , together with @xmath429 , form a togliatti system @xmath568 of degree @xmath84 .
we distinguish the following possibilities : * at least one of the monomials @xmath569 is the @xmath343th power of a variable , so @xmath570 ; or * @xmath571 .
+ in case ( i ) , if @xmath572 , @xmath568 is trivial , which implies that @xmath47 contains a trivial togliatti system and therefore is non minimal : contradiction . if @xmath258 , @xmath573 and @xmath574 .
+ in case ( ii ) , we can apply the above argument to @xmath568 , and so on , by induction .
+ in any case , applying repeatedly this procedure , possibly involving different variables , we arrive to a togliatti system @xmath568 of degree @xmath439 with @xmath575 , which is obtained from @xmath47 dividing the monomials @xmath564 by a common monomial factor @xmath576 .
if @xmath25 , we conclude with the help of macaulay2 .
if @xmath26 , by lemma [ d=3,2n+3 ] , @xmath568 is trivial of type @xmath577 or @xmath578 . in both cases @xmath47 is not minimal and we are done . if @xmath25 and @xmath258 one can check with the help of macaulay2 that there exist two types of minimal togliatti systems @xmath47 with @xmath579 , both non smooth , precisely @xmath580 and @xmath581 . we note that if @xmath438 the ideal @xmath582 , with @xmath583 , belongs to @xmath584 , while if @xmath439 then @xmath585 for any @xmath26 .
computations made with macaulay2 illustrate the complexity of the general case .
however , some ranges and some sporadic values can be covered .
for example : for any @xmath586 and for any @xmath68 , @xmath587 , there exists @xmath588 with @xmath589 .
( notice that when @xmath25 we have @xmath590 ) .
in fact , it is enough to take @xmath591 where @xmath592 and @xmath593 is a monomial of degree @xmath594 containing only @xmath595 .
nevertheless if we delete the smoothness hypothesis , we can generalize proposition [ interval ] and we get [ interval2 ] with the above notation we have : * for any @xmath150 , @xmath596 . * for any @xmath150 , @xmath597 . * for any @xmath150 , @xmath25 and any integer @xmath68 with @xmath598 , there exists @xmath599 with @xmath473 .
\(i ) it follows from theorem [ mainthm1 ] .
\(ii ) by definition we have @xmath600 for any @xmath150 .
let us prove that @xmath601 , i.e. there exists @xmath602 with @xmath603 . consider @xmath604 we have @xmath605 + d-2 \\ & = & d+1+\sum _ { i=2}^{n-1}{d-1+i\choose i } \\ & = & \sum _ { i=0}^{n-1}{d-1+i\choose i}\\ & = & { d-1+n\choose n-1}.\end{array}\ ] ] when we substitute @xmath242 by @xmath606 the @xmath94 generators of @xmath47 become @xmath33-linearly dependent ; so @xmath47 fails wlp in degree @xmath84 ( theorem [ teathm ] ) and @xmath47 is minimal because no proper subset of the generators of @xmath47 defines a togliatti system .
therefore , @xmath607 .
\(iii ) assume @xmath25 . for @xmath608
we take @xmath609 , for @xmath610 we take @xmath611 and for @xmath612 we take @xmath613
. we will now proceed by induction on @xmath0 . in the case
@xmath258 we exhibit an explicit example for any @xmath614 ( note that the case @xmath615 is covered by the example given in ( ii ) ) : * @xmath616 : @xmath617 ( smooth ) ; * @xmath618 : @xmath619 ; * @xmath620 : @xmath621 ; * @xmath622 : @xmath623 ; * if @xmath624 : the systems described in remark [ trivialb ] work in this case .
we suppose now @xmath572 and we will prove that for any @xmath625 there exists @xmath626 with @xmath473 .
indeed , for any @xmath627 we take @xmath628 with @xmath629 and we define @xmath630 .
note that @xmath631 and @xmath632 .
observe also that @xmath633 and @xmath634 .
so , it only remains to cover the values of @xmath68 , @xmath635 to this end , for any @xmath636 we define @xmath637 first of all we observe that @xmath638
. therefore , when @xmath639 ranges from @xmath640 to @xmath84 we sweep the interval @xmath641 $ ] . by proposition [ failurewlp ] to prove that @xmath642 it is enough to show that there is a surface @xmath224 of degree @xmath84 containing all integral points of @xmath643 .
since @xmath644 , we have @xmath645 where @xmath646 is a surface of degree @xmath647 containing all integral points of @xmath648 . the surfaces
@xmath646 of degree @xmath647 are parametrized by a @xmath33-vector space of dimension @xmath649 .
on the other hand , to contain the aligned @xmath84 points of @xmath650 imposes @xmath367 conditions on the surfaces of degree @xmath647 , to contain the points of @xmath651 imposes @xmath652 conditions , respectively , and finally to contain the points of @xmath653 imposes @xmath654 conditions .
summing up we have @xmath655 conditions .
therefore , there exists at least a surface @xmath646 of degree @xmath647 through all integral points of @xmath656 and , hence a surface @xmath645 of degree @xmath84 containing all integral points of @xmath643 . for @xmath657 , with macaulay2
we have obtained the list of all minimal togliatti systems with @xmath658 .
the computations become too heavy for @xmath659 .
in this section we restrict our attention to the case @xmath4 and we will analyze whether the syzygy bundle @xmath660 on @xmath13 associated to a minimal smooth monomial togliatti system @xmath661 is @xmath662-(semi)stable .
a syzygy bundle @xmath663 on @xmath36 is a rank @xmath664 vector bundle defined as the kernel of an epimorphism @xmath665 where @xmath666 $ ] is an artinian ideal , and @xmath667 . when @xmath668 , we write @xmath669 instead of @xmath663 .
let e be a vector bundle on @xmath36 and set @xmath670 the vector bundle @xmath671 is said to be @xmath672-semistable in the sense of mumford - takemoto if @xmath673 for all non - zero subsheaves @xmath674 with @xmath675 ; if strict inequality holds then @xmath671 is @xmath662-stable .
using klyachko results on toric bundles ( @xcite , @xcite and @xcite ) , brenner deduced the following nice combinatoric criterion for the ( semi)stability of the syzygy bundle @xmath677 in the case where the associated forms @xmath678 are all monomials .
indeed , we have [ combinatoria ] let @xmath679 $ ] be a monomial artinian ideal .
set @xmath680 .
then the syzygy bundle @xmath677 on @xmath36 associated to @xmath47 , is @xmath672-semistable ( resp .
@xmath672-stable ) if and only if for every @xmath681 , @xmath41 , the inequality @xmath682 holds , where @xmath683 is the degree of the greatest common factor of the monomials @xmath684 .
\(1 ) if we consider the monomial artinian ideal @xmath685 $ ] , inequality ( [ ineq1 ] ) is strictly fulfilled for any proper subset @xmath686 .
therefore the syzygy bundle @xmath671 associated to @xmath47 is @xmath662-stable .
\(2 ) if we consider the monomial artinian ideal @xmath687 $ ] , then for the subset @xmath688 inequality ( [ ineq1 ] ) is not fulfilled .
therefore the syzygy bundle @xmath660 associated to @xmath47 is not @xmath672-stable .
in fact , the slope of @xmath660 is @xmath689 and the syzygy sheaf @xmath166 associated to @xmath690 is a subsheaf of @xmath660 with slope @xmath691 . since @xmath692 , we conclude that @xmath671 is not @xmath672-stable .
[ rem ] let @xmath47 be a monomial artinian ideal generated by @xmath68 monomials @xmath693 of degree @xmath0 .
it easily follows from the above proposition that the syzygy bundle @xmath669 on @xmath36 associated to @xmath47 is @xmath672-(semi)stable if and only if for every subset @xmath694 with @xmath695 , @xmath696 where @xmath683 is the degree of the greatest common factor of the monomials in @xmath690 .
* @xmath660 is @xmath672-stable if and only if , up to a permutation of the coordinates , one of the following cases holds : * * @xmath290 , @xmath186 and @xmath698 .
* * @xmath699 , @xmath356 and @xmath700 or @xmath701 @xmath702 or @xmath703 .
* @xmath660 is properly @xmath672-semistable if and only if , up to a permutation of the coordinates , one of the following cases holds : * * @xmath704 , @xmath186 and @xmath705 . * * @xmath704 , @xmath186 and @xmath706 or @xmath707 @xmath708 . * in all other cases , @xmath660 is unstable .
to prove ( 2 ) we check that inequality ( [ ineq3 ] ) is satisfied for any proper subset @xmath716 , @xmath714 , with @xmath718 and there is a subset @xmath719 , @xmath714 , with @xmath720 and verifying @xmath721 .
for instance , for @xmath722 it is enough to take @xmath723 since @xmath724 .
\(3 ) finally let us check that the syzygy bundle @xmath660 associated to trivial togliatti systems @xmath725 are always @xmath672-unstable .
note that @xmath435 is a monomial of degree @xmath84 and @xmath726 , @xmath727 , are monomials of degree @xmath0 .
for the subset @xmath728 inequality ( [ ineq3 ] ) becomes @xmath729 and @xmath660 is @xmath672-unstable .
indeed , the slope of @xmath660 is @xmath730 and the syzygy sheaf @xmath166 associated to @xmath690 is a subsheaf of @xmath660 with slope @xmath731 .
therefore , @xmath732 and we conclude that @xmath660 is @xmath672-unstable .
r. albini , _ sistemi di togliatti _ , tesi di laura magistrale in geometria algebrica , universit degli studi di trieste , 2013
. h. brenner and a. kaid , _ syzygy bundles on @xmath733 and the weak lefschetz property _ , illinois j. math .
* 51 * ( 2007 ) , 12991308 .
j. watanabe , _ the dilworth number of artinian rings and finite posets with rank function _
, commutative algebra and combinatorics , advanced studies in pure math .
vol . 11 , kinokuniya co. north holland , amsterdam ( 1987 ) , 303312 . | we compute the minimal and the maximal bound on the number of generators of a minimal smooth monomial togliatti system of forms of degree @xmath0 in @xmath1 variables , for any @xmath2 and @xmath3 .
we classify the togliatti systems with number of generators reaching the lower bound or close to the lower bound .
we then prove that if @xmath4 ( resp @xmath5 ) all range between the lower and upper bound is covered , while if @xmath6 ( resp .
@xmath7 ) there are gaps if we only consider smooth minimal togliatti systems ( resp .
if we avoid the smoothness hypothesis ) .
we finally analyze for @xmath4 the mumford - takemoto stability of the syzygy bundle associated to smooth monomial togliatti systems . |
the pauli exclusion principle ( pep ) is a consequence of the spin - statistics connection @xcite and plays a fundamental role in our understanding of many physical and chemical phenomena , from the periodic table of elements , to the electric conductivity in metals , to the degeneracy pressure , which makes white dwarfs and neutron stars stable , just to cite few ones .
although the principle has been spectacularly confirmed by the number and accuracy of its predictions , its foundation lies deep in the structure of quantum field theory and has defied all attempts to produce a simple proof , as nicely stressed by r. feynman @xcite .
given its basic standing in quantum theory , it seems appropriate to carry out precise tests of the pep validity and , indeed , in the last fifty years , several experiments have been performed to search for possible small violations @xcite .
often , these experiments were born as by - products of experiments with a different objective ( e.g. dark matter searches , proton decay , etc .. ) , and most of the recent limits on the validity of pep have been obtained for nuclei or nucleons .
concerning the violation of pep for electrons , greenberg and mohapatra @xcite examined all experimental data which could be related , directly or indirectly , to pep , up to 1987 . in their analysis
they concluded that the probability that a new electron added to an antisymmetric collection of n electrons might form a mixed symmetry state rather than a totally antisymmetric state is @xmath2 . in 1988 , ramberg and snow @xcite drastically improved this limit with a dedicated experiment , searching for anomalous x - ray transitions , that would point to a small violation of pep in a copper conductor .
the result of the experiment was a probability @xmath3 that a new electron circulating in the conductor would form a mixed symmetry state with the already present copper electrons .
we have set up an improved version of the ramberg and snow experiment , with a higher sensitivity apparatus @xcite .
our final aim is to lower the pep violation limit for electrons by at least 4 orders of magnitude , by using high resolution charge - coupled devices ( ccd ) as soft x - rays detectors @xcite , and decreasing the effect of background by a careful choice of the materials and sheltering the apparatus in an underground laboratory . in the next sections we describe the experimental setup ,
the outcome of a preliminary measurement performed in the frascati national laboratories ( lnf ) of infn in 2005 , along with a brief discussion on the results and the foreseen future improvements in the gran sasso national laboratory ( lngs ) of infn .
the idea of the vip ( violation of the pauli exclusion principle ) experiment was originated by the availability of the dear ( da@xmath4ne exotic atom research ) setup , after it had successfully completed its program at the da@xmath4ne collider at lnf - infn @xcite .
dear used charge - coupled devices ( ccd ) as detectors in order to measure exotic atoms ( kaonic nitrogen and kaonic hydrogen ) x - ray transitions .
ccd s are almost ideal detectors for x - rays measurement , due to their excellent background rejection capability , based on pattern recognition , and to their good energy resolution ( 320 ev fwhm at 8 kev in the present measurement )
. the experimental method , originally described in @xcite , consists in the introduction of new electrons into a copper strip , by circulating a current , and in the search for x rays resulting from the @xmath5 anomalous radiative transition that occurs if one of the new electrons is captured by a copper atom and cascades down to the 1s state already filled by two electrons of opposite spin . the energy of this transition would differ from the normal k@xmath6 transition by about 300 ev ( 7.729 kev instead of 8.040 kev ) @xcite , providing an unambiguous signal of the pep violation .
the measurement alternates periods without current in the copper strip , in order to evaluate the x - ray background in conditions where no pep violating transitions are expected to occur , with periods in which current flows in the conductor , thus providing `` fresh '' electrons , which might possibly violate pep .
the fact that no pep violating transitions are expected to be present in the measurement without current is related to the consideration that any initial conduction electron in the copper that was in a mixed symmetry state with respect to the other copper electrons , would have already cascaded down to the @xmath7 state and would therefore be irrelevant for the present experiment .
the rather straightforward analysis consists in the evaluation of the statistical significance of the normalized subtraction of the two spectra , with and without current , in the energy region where the pep violating transition is expected .
the vip setup consists of a copper cylinder , 4.5 cm in radius , 50 @xmath8 m thick , 8.8 cm high , surrounded by 16 equally spaced ccd s @xcite .
the ccd s are at a distance of 2.3 cm from the copper cylinder , grouped in units of two chips vertically positioned .
the setup is shown in fig .
the chamber is kept at high vacuum to minimize x - ray absorption and to avoid condensation on the cold surfaces . the copper target ( the copper strip where the current flows and new electrons are injected from the power supply ) is at the bottom of the setup .
the ccd s surround the target and are supported by cooling fingers that start from the cooling heads in the upper part of the chamber .
the ccd readout electronics is just behind the cooling fingers ; the signals are sent to amplifiers on the top of the chamber .
the amplified signals are read out by adc boards in a data acquisition computer .
more details on the ccd-55 performance , as well as on the analysis method used to reject background , can be found in @xcite the measurements reported in this paper have been performed in the period 21 november - 13 december 2005 , at the frascati national laboratories of infn , italy .
two types of measurements were performed : * 14510 minutes ( about 10 days ) of measurements with a 40 a current circulating in the copper target ; * 14510 minutes of measurements without circulating current , where ccd s were read - out every 10 minutes . the two resulting x - ray spectra are shown in figure 2 a ) , with circulating current , and b ) , without current .
the spectra refer to 14 ccd s ( out of 16 ) , due to noise problems in the remaining 2 .
in order to obtain the number of x - rays due to the possible pep violating transitions , the spectrum without current was subtracted from the one with current .
the resulting subtracted spectrum is shown in figure 3 a ) ( whole energy scale ) and b ) ( a zoom on the region of interest ) .
the region of interest , from 7.564 to 7.894 kev , is defined by the ccd energy resolution ( 320 ev fwhm ) at the @xmath9 copper transition ( 8.04 kev ) , with an additional uncertainty of 10 ev , to account for the theoretical uncertainty in the calculation of the pep violating transition energy .
the numbers of x rays in the region of interest were : * at i=40 a : n@xmath10 ; * for i=0 a : n@xmath11 ; * for the subtracted spectrum : @xmath12 .
[ p18kev.eps ] for the parametrization of the results in a pauli principle violating theory , we use the notation of ignatiev and kuzmin @xcite , which has been incorporated in the paper of greenberg and mohapatra @xcite : even though the model of ignatiev and kuzmin has been later shown to be incompatible with quantum field theory @xcite , the parameter @xmath13 that measures the degree of pep violation has stuck and is still found in the literature , also because it is easy to show that it is related to the parameter @xmath14 of quon theory , by the relation : @xmath15 @xcite ( in quon theory , @xmath16 , where @xmath17 corresponds to fermions and @xmath18 corresponds to bosons , so that here @xmath14 must be close to -1 and @xmath19 must be very small , because we are dealing with electrons ) .
moreover , we used this parametrization for an easy comparison of our results with the previous ramberg and snow ones @xcite , since the same has been used in that paper . in @xcite a pair of electrons in a mixed symmetry state has the probability @xmath20 for the symmetric component and @xmath21 for the usual antysymmetric one .
the parameter @xmath22 is related , then , to the probability that an electron violates pep ( see also @xcite for further details ) . to determine the experimental limit on @xmath22 from our data
, we used the same arguments of ramberg and snow , to compare the results .
the number of electrons that pass through the conductor , which are new for this conductor , is : @xmath23 where @xmath24 is the electron electric charge , @xmath25 is the current intensity and @xmath26 represents the time duration of the measurement with current on . each new electron will undergo a large number of scattering processes on the atoms of the copper lattice .
the minimum number of these internal scattering processes per electron , defined as @xmath27 , is of order @xmath28 , where @xmath29 is the length of the copper electrode ( 8.8 cm in our case ) and @xmath8 is the mean free path of electrons in copper .
the latter parameter is obtained from the resistivity of the metal .
we assume that the capture probability ( aside from the factor @xmath30/2 ) is greater than @xmath31 of the scattering probability . the acceptance of the 14 ccd detectors and the probability that an x ray of about 7.6 kev , the energy of the possible anomalous transition generated in the copper target , is not absorbed inside the copper itself , were evaluated by a monte carlo simulation of the vip setup , based on geant 3.21 .
this probability turns out to be 2.1% .
moreover , a ccd efficiency equal to 48% for a 7.6 kev x ray was considered .
all these factors built up the so called @xmath32 ( @xmath33% ) .
the number of x rays generated in the pep violating transition , @xmath34 , is then related to the @xmath20 parameter by : @xmath35 then , for @xmath36 c , @xmath37 cm , @xmath38 cm , @xmath39 c , we get @xmath40 the difference of events between the measurements with and without current , reported in the previous section , is @xmath41 .
taking as a limit of observation three standard deviations , we get for the pep violating parameter : @xmath42 we can interpret this as a limit on the probability of pep violating interactions between external electrons and copper atoms : @xmath1 . we have thus improved the limit obtained by ramberg and snow by a factor about 40 .
the paper reports a new measurement of the pep violation limit for electrons , performed by the vip collaboration at lnf - infn .
the search of a tiny violation was based on a measurement of pep violating x - ray transitions in copper , under a circulating 40 a current .
a new limit for the pep violation for electrons was found : @xmath1 , lowering by almost two orders of magnitude the previous one @xcite .
we shall soon repeat the measurement in the gran sasso
infn underground laboratory , at higher integrated currents . from preliminary tests
, it appears that the x - ray background in the lngs environment is a factor 10 - 100 lower than in the frascati laboratories .
a vip measurement of two years ( one with current , one without current ) at lngs , to start in spring 2006 , will then bring the limit on pep violation for electrons into the 10@xmath43 - 10@xmath44 region , which is of particular interest @xcite for all those theories related to possible pep violation coming from new physics .
w. pauli , phys .
* 58 * ( 1940 ) 716 .
r. p. feynman , r. b. leighton , and m. sands : `` the feynman lectures on physics '' , vol .
3 , ( addison - wesley , reading , ma , 1963 ) .
r. bernabei _ et al .
_ , phys . lett . *
b408 * ( 1997 ) 439 .
h. o. back _ et al .
_ ( borexino collaboration ) , eur .
j. * c37 * ( 2004 ) 421 .
r. c. hilborn and c. l. yuca , phys .
* 76 * ( 1996 ) 2844 .
nemo collaboration , nucl .
b87 * ( proc . suppl . ) ( 2000 ) 510 .
et al . _ , j. phys .
g : nucl . part .
* 17 * ( 1991 ) s355 .
m. tsipenyuk , a. s. barabash , v. n. kornoukhov , and b. a. chapyzhnikov , radiat .
phys . chem .
* 51 * ( 1998 ) 507 .
greenberg and r.n .
mohapatra , phys .
* 59 * ( 1987 ) 2507 .
e. ramberg and g.a .
snow , phys .
b238 * ( 1990 ) 438 . the vip proposal , lnf - lngs proposal , september 2004 ( http://www.lnf.infn.it/esperimenti/vip ) .
see e.g. j.l .
culhane , nucl .
instr . and meth .
* a310 * ( 1991 ) 1 ; j .- p .
egger , d. chatellard , e. jeannet , particle world * 3 * ( 1993 ) 139 ; g. fiorucci _ et al .
instr . and meth .
* a292 * ( 1990 ) 141 ; d. varidel _
et al . _ ,
instr . and meth .
* a292 * ( 1990 ) 147 ; r.p .
et al . _ ,
instr . and meth .
* a372 * ( 1995 ) 372 .
t. ishiwatari _
_ , phys . lett .
* b593 * ( 2004 ) 48 ; g. beer _ et al .
* 94 * ( 2005 ) 212302
. s. di matteo and l. sperandio , vip note , ir-04 , april 26 , 2006 ( the energy shift has been computed by p. indelicato - private communication ) .
ccd-55 from eev ( english electric valve ) , waterhouse lane , chelmsford , essex , cm1 2qu , uk .
t. ishiwatari _
et al . _ ,
instrum . and meth .
research * a556 * ( 2006 ) 509 .
ignatiev and v. a. kuzmin , yad . fiz .
* 46 * ( 1987 ) 786 ; ictp preprint ic/87/13 ( 1987 ) ; a. yu ignatiev , arxiv : hep - ph/0509258 .
a. b. govorkov , phys .
lett * a137 * ( 1989 ) 7 .
greenberg , phys .
* d43 * ( 1991 ) 4111 .
okun , comments nucl .
( 1989 ) 998 .
i. duck and e.c.g .
sudarshan , am .
j. of physics * 66 * ( 1998 ) 284 . | the pauli exclusion principle ( pep ) is one of the basic principles of modern physics and , even if there are no compelling reasons to doubt its validity , it is still debated today because an intuitive , elementary explanation is still missing , and because of its unique stand among the basic symmetries of physics .
the present paper reports a new limit on the probability that pep is violated by electrons , in a search for a shifted k@xmath0 line in copper : the presence of this line in the soft x - ray copper fluorescence would signal a transition to a ground state already occupied by 2 electrons .
the obtained value , @xmath1 , improves the existing limit by almost two orders of magnitude . , , , , , , , , , , , , , , , , , , ,
, symmetrization principle , identical particles , tests of quantum field theories , anomalous atomic transitions , x rays , ccd 11.30.-j ; 03.65.-w ; 29.30.kv ; 32.30.rj |
a billiard ball , i.e. a point mass , moves inside a polygon @xmath0 with unit speed along a straight line until it reaches the boundary @xmath1 , then instantaneously changes direction according to the mirror law : `` the angle of incidence is equal to the angle of reflection , '' and continues along the new line .
if the trajectory hits a corner of the polygon , in general it does not have a unique continuation and thus by definition it stops there .
billiards in polygons are easy to describe , but it is difficult to prove deep theorems about them because of a lack of machinery .
for example , it is unknown if every polygon contains a periodic billiard orbit . on the other hand , for so called rational polygons
, one can apply teichmller theory to obtain many deep theorems .
it is known that all rational polygons possess many periodic orbits @xcite , and in fact they are dense in the phase space @xcite .
galperin , stepin and vorobets proved many other interesting results about periodic orbits @xcite .
many of these results can be found in the introductory book by tabachnikov @xcite , and in several survey articles @xcite .
the main result of this article produces the first non rational polygons for which the above mentioned result is true : _ periodic billiard trajectories are dense in the phase space for an open set of right triangles _
( theorem [ thm0 ] ) .
this result is proved by a careful analysis of the symmetries of perpendicular periodic orbits ( theorem [ thm1 ] ) . using these symmetries
we prove that _ there is a unique perpendicular escape orbit _
( theorem [ thm2 ] ) and that _ the invariant surface which contains the perpendicular direction is completely foliated by perpendicular periodic orbits _ ( theorem [ pascal ] ) .
the billiard in a right triangle is well known to be equivalent to the mechanical system of two elastic point particles in an interval ( see for example @xcite ) , thus our results hold for this system as well .
cipra , hanson and kolan have shown that almost every orbit which is perpendicular to the base of a right triangle is periodic @xcite . here
the almost every is with respect to the length measure on the side of the triangle in question .
periodic billiard orbits always come in _ strips _ , i.e. for any @xmath2 whose billiard orbit is periodic where @xmath3 and @xmath4 is any inward pointing direction there is an open interval @xmath5 such that @xmath6 and for any @xmath7 the orbit of @xmath8 visits the same sequence of sides as @xmath9 and thus in particular is periodic ( see figure 1 ) .
a maximal width strip will be called a _
beam_. all orbits in a periodic beam have the same period except perhaps the middle orbit which has half the period in the case its period is odd .
the results of @xcite imply that a set of full measure of the base of the triangle is covered by an at most countable union of intervals such that each of these intervals forms a periodic beam .
they mention `` our computational evidence also suggest that the trajectory at the middle of each ( perpendicular periodic ) beam hits the right - angle vertex of the triangle . ''
this had also been noticed in an earlier article by ruijgrok @xcite .
they also speculate that there is at most one nonsingular , non periodic trajectory .
these speculations were the starting point of this research , for this article .
i present an elementary proof of these facts for irrational right triangles with smaller angle @xmath10 satisfying @xmath11 .
make the _ convention _ that one leg of the right triangle is horizontal with the smaller angle @xmath10 of the right triangle being the angle between this leg and the hypotenuse , and that the word perpendicular ( orbit , beam , etc . )
refers to perpendicularity to the horizontal leg .
any billiard trajectory which hits a right - angled vertex ( or more generally a vertex with angle @xmath12 for some positive integer @xmath13 ) has a unique continuation .
reflect the triangle in the sides of right angle to obtain a rhombus .
the study of the billiard in the triangle reduces to that in the rhombus ( see next section for details ) . throughout the article
all beams considered will be with respect to the angle coding in the rhombus defined in the next section .
[ thm1 ] for any irrational right triangle whose smaller angle satisfies @xmath11 consider any perpendicular periodic beam of period @xmath14
. then 1 .
the midpoint of the beam hits the right - angle vertex of the triangle ( i.e. the mid point of the rhombus ) , 2 .
the beam returns to itself after half its period with the opposite orientation , and 3 .
@xmath15 is even and the first @xmath16 letters of the code of the beam form a palindrome .
the theorem implies that when viewed as an object in the phase space of the billiard flow in the triangle the beam is a mebius band .
this is not in contradiction with the well known construction , of invariant surfaces ( see for example @xcite in the rational case and @xcite in the irrational case ) since the directional billiard flow is isomorphic to the geodesic flow on the invariant surface for any direction except for directions which are parallel to a side of the polygon .
the direction we are considering is such a direction .
the symmetry of the beam is reminiscent of the symmetry of the beam of periodic orbits around a periodic orbit of odd period mentioned above .
galperin , stepin and vorobets showed that any perpendicular orbit whose period is not a multiple of 4 is unstable under perturbation @xcite .
viewed as an orbit in the rhombus the period is a multiple of 4 , however a simple argument shows that the period is not a multiple of 4 when the orbit is viewed as an orbit in a right triangle @xcite .
call an orbit _ recurrent _ if its code @xmath17 satisfies @xmath18 for some @xmath19 . call a nonrecurrent orbit an _ positive escape orbit _ if @xmath20 and _ negative escape orbit _ if @xmath21 .
call a direction _ simple _ if there are no generalized diagonals in the invariant surface containing the direction . here
the invariant surface is thought of as arising from the right triangle , so that the vertex at the right angle is considered as a singularity .
[ thm2 ] @xmath22 1 .
consider an irrational right triangle . in any simple direction
then there is at most one non - singular positive escape orbit and at most one non - singular negative escape orbit .
2 . for any irrational right triangle
whose smaller angle satisfies @xmath11 , there is at most one non - singular escape orbit in the perpendicular direction .
if it exists it is both a positive and negative escape orbit .
if we do not care if the orbit is singular or not , then we can remove the words at most from the statement of the theorem .
theorem [ thm2 ] is a substantial strengthening of a particular case of a result of @xcite who show that the set of nonrecurrent orbits has measure zero and of a particular case of a result of @xcite who show that this set has lower box counting dimension at most one half .
boshernitzan has conjectured that given a rational triangle , every nonsingular orbit is periodic in the invariant surface containing a perpendicular direction @xcite .
i prove the following analog for irrational right triangles .
[ pascal ] fix an irrational right triangle whose smaller angle satisfies @xmath11 .
consider the invariant surface of @xmath23 in the perpendicular direction .
then all the nonsingular orbits on @xmath23 except the unique escape orbit are periodic . combining theorem [ pascal ] with the main theorem of @xcite yields [ thm0 ] periodic orbits
are dense in the phase space of irrational right triangles whose smaller angle satisfies @xmath11 .
there is a nice introductory book on billiards by tabachnikov @xcite and several survey articles @xcite which can be consulted for details on polygonal billiards in general . for a rational polygon
there is a well known construction of invariant surfaces , they are always compact .
they same construction leads to noncompact invariant surfaces for irrational polygons .
the invariant surfaces for an irrational right triangle can be thought of as having a @xmath24-quasiperiodicity @xcite .
the proof is based on the procedure of _ unfolding of a billiard trajectory_. instead of reflecting the trajectory with respect to a side of a polygon reflect the polygon in this side .
thus the trajectory is straightened to a line with a number of isometric copies of the polygon skewered on it ( figure 2a ) .
fix an orbit segment .
there is a _ strip _ around this trajectory segment such that the same sequence of reflections is made by all trajectories in the strip ( figure 2b ) .
the number of reflections is called the _ length _ of the strip . call a maximal width strip a _
beam_. the boundary of a periodic beam consists of one or more trajectory segments which hit a vertex of the polygon .
if this vertex is the right - angle one , then the sequence of reflections on both sides of it is essentially the same since the singularity due to such a vertex is removable .
this enables us to consider the billiard inside a rhombus which consists of four copies of the right triangle and unfold the rhombus ( figure 3 ) .
the central and reflectional symmetries of the rhombus play an essential part in the proofs of the theorems . throughout the article unfoldings
will be of the billiard in the rhombus and thus billiard orbits through its center will be considered as defined . as we follow the reflections along a straight line trajectory we see that each flip rotates the rhombus by @xmath25 where @xmath10 is one of the interior angles of the rhombus .
thus we can _ label _ the rhombi with integers according to the total number of ( clockwise ) rotations by @xmath26 .
each orbit is _ coded _ by the sequence of labelled rhombi it visits , for example any orbit in the beam @xmath27 in figure 3 has code 0121210 .
it is often convenient to consider labelled rhombus in a projective sense as an interval , namely for rhombus @xmath28 consider the beam of length 1 with code @xmath28 and then the associated interval @xmath29 is just the perpendicular width of of the beam ( see figure [ split ] ) .
the @xmath28th - rhombus or the associated interval @xmath29 will often be referred to as _ level _ @xmath28 .
all reference to the length of a beam will pertain to the number of rhombi it crosses which differs from its length in the right triangle .
let us begin with the following fact .
[ lemma1 ] suppose the @xmath27 and @xmath30 are parallel beams ( not necessarily perpendicular or periodic ) with the same codes , then @xmath31 .
the proof is by induction .
fix an initial direction .
the base case is clear , there is a single beam with code @xmath32 .
furthermore , there is a single beam with code @xmath33 , a single beam with code @xmath34 and these two beams of length two are separated by an orbit which hits a vertex of the rhombus .
assume that every beam with length @xmath13 is the unique beam with its code .
fix a beam of length @xmath13 and denote by @xmath35 the last rhombus it visits .
the @xmath35th rhombus is attached to the @xmath36st by two parallel sides and to the @xmath37st along the other two parallel sides .
there are two points which are common to the @xmath37st , @xmath35th and @xmath36st rhombi . during a given pass through the @xmath35th rhombus ,
the beam can hit only one of the two in exiting the @xmath35th rhombus ( it can hit the other in entering the @xmath35 rhombus ) .
the orbit which splits a beam is emphasized in figure [ split ] .
if the beam does not arrive at this point then it does not split and the beam can be continued to a unique beam of length @xmath38 .
if the beam reaches this point then its continuation is split into two sub - beams of length @xmath38 with the resulting sub - beams code differing in the @xmath39st place .
next we need the following fact .
[ lemma2 ] the three statements in theorem [ thm1 ] are equivalent .
perpendicular periodic beams must have even period since they must retrace their orbits between the two perpendicular hits .
fix a perpendicular periodic beam of period @xmath40 consider the beam of length @xmath15 between the two perpendicular collisions ( see figure 3 ) .
@xmath41 and @xmath42 suppose that the midpoint of this beam hits the right - angle vertex .
the central symmetry of the rhombus at this collision point implies that the beam itself is centrally symmetric in this point and thus has palindromic code .
furthermore @xmath15 is even since the symmetry is about an interior point of the rhombus .
label by @xmath43 the interval of the departure and by @xmath44 the interval of the arrival .
both intervals are contained in the rhombus labelled 0 , the central symmetry of the beam implies that @xmath43 and @xmath44 are centrally symmetric .
@xmath45 by assumption the initial interval @xmath43 of the beam and the final interval @xmath44 fold to the same interval in the triangle thus the midpoint of @xmath43 must have period @xmath15 or have a singular orbit as an orbit in the triangle .
it is easy to see that the orbit can not be periodic , since any perpendicular periodic orbit must be twice perpendicular and the collision at time @xmath46 is not perpendicular .
thus the orbit is singular in the triangle . since it was nonsingular in the rhombus it hit the right - angled vertex of the triangle .
@xmath47 consider the central symmetry of the palindromic beam around its center @xmath48 .
this takes the beam to a priori another beam with the same code .
lemma [ lemma1 ] implies that the beams must coincide .
thus the beam is centrally symmetric about its center .
since @xmath15 is even the point @xmath48 must lie in the interior of the rhombus visited by the strip at step @xmath46 .
however , viewed in the rhombus one sees that the beam can only be centrally symmetric around the center of the rhombus and no other point , thus @xmath48 must be the center of the rhombus .
we now turn to a detailed analysis of beams which start in a given direction , and either return to that direction or another fixed direction .
[ lemma5 ] fix an angle @xmath49 and consider the angle coding of the billiard with @xmath49 taken as level @xmath32 .
suppose @xmath50 .
consider the set @xmath51 of points whose beam of length 2 has code @xmath33 .
this set is an interval , partition it into intervals such that for each partition interval there exists @xmath52 such that the initial orbit code @xmath53 of all points in the interval does not depend on the points and is such that @xmath54 , @xmath55 for @xmath56 and @xmath57 is equal to either @xmath32 or @xmath58 .
then there are at most @xmath59 such codes and corresponding subintervals of @xmath51 .
the interiors of these subintervals are pairwise disjoint and their union covers @xmath51 .
this lemma has been essentially proven in @xcite and somewhat more explicitly in @xcite .
i repeat the proof here for completeness .
consider the beam of length two whose initial code is @xmath33 .
follow all the orbits in this beam until they reach level @xmath58 or return to level @xmath32 at some time @xmath15 . for the beam to split
, an orbit in the beam must reach a vertex .
this can be seen for example in figure 3 , the beam ( not drawn in the picture ) just to the left of the beam @xmath27 has code @xmath60 ( the copy of the rhombus corresponding to last @xmath32 is not drawn in the picture ) .
these two beams are split by a vertex in ( the @xmath61nd occurence ) of the @xmath62st rhombus . for each @xmath63
there is a single vertex @xmath64 which can split the beam in rhombus @xmath63 .
since the billiard is invertible this vertex can only be reached from at most one point in the @xmath32th rhombus ( without returning to the rhombus ) .
the set @xmath65 of vertices has cardinality @xmath66 , therefore there are at most @xmath58 distinct beams and thus codes .
[ lemma3 ] fix an angle @xmath49 and consider the angle coding of the billiard with @xmath49 taken as level @xmath32 . for any @xmath67 we can partition level @xmath32 into a finite number of intervals with each interval in one of three classes , @xmath68 such that the orbits of any point in an interval in class @xmath69 and @xmath70 does not reach level @xmath23 or @xmath58 . furthermore 1 .
the forward orbit of each boundary point of each interval is singular , 2 .
the interior of each of the intervals in @xmath69 consists periodic points and the boundaries of each periodic family consists of generalized diagonals , 3 .
the intervals in @xmath70 consists of one ( or several ) minimal interval exchange transformations ( i.e. on a finite number of intervals ! ) and the boundary of each minimal component consist of generalized diagonals and 4 .
the orbit of each point in the interior of an interval in @xmath71 reaches level @xmath23 or @xmath58 .
we have the following immediate corollary .
[ cor1 ] the orbit of every point in level @xmath32 which does not reach level @xmath23 or @xmath58 is either singular or returns to level @xmath32 .
the classes may well be empty , for example if we choose a direction @xmath49 for which the surface has no generalized diagonals then both @xmath69 and @xmath70 must be empty .
galperin has shown that for almost all right triangles there exists a @xmath49 such that the set @xmath70 is nonempty @xcite .
consider the cover of @xmath51 defined in the previous lemma and use the subintervals which return to define a partially defined iet on level @xmath32 .
do the same for points whose code starts with @xmath72 .
call the domain of definition @xmath73 .
do the same for the backwards dynamics and remark that these two partially defined maps are inverses of each other where ever they are defined .
the total length of the intervals where the forward map is not defined is equal to the total length of the intervals where the backwards map is not define .
thus we can ( in an arbitrary manner ) complete the definition of the partially defined map to an iet , which we call the ghost map .
the ghost iet agrees with the partially defined first return billiard map whenever it was defined . since the ghost map is an iet with interval of definition being level @xmath32 .
the well known topological decomposition holds , the interval of definition is decomposed into periodic and minimal components , with the boundary of the components consisting of generalized diagonals ( see for example @xcite ) .
any point whose orbit enters an interval on which the ghost dynamics differs from the true dynamics will reach level @xmath23 or @xmath58 following the true dynamics . on the other hand the topological decomposition mentioned above can be applied to the point for which the ghost and true dynamics always agree .
we remark that a ghost minimal component will , by minimality be either completely contained in @xmath73 or every orbit will enter an interval on which the ghost dynamics differs from the true dynamics and thus will reach level @xmath23 or @xmath58 following the true dynamics .
we next begin the analysis of simple directions .
the main reason to analyze such directions is as a `` warm - up '' for perpendicular directions .
[ lemma4 ] suppose @xmath49 is simple .
then for all integers @xmath74 1 .
there exists a nonsingular orbit segment starting in level @xmath23 and ending in level @xmath58 and 2 . a nonsingular orbit segment starting in level @xmath58 and ending in level @xmath23 .
we first remark that if there is a nonsingular orbit from @xmath23 to @xmath58 then since the map is locally an isometry there is a whole interval which maps from @xmath23 to @xmath58 . since a.e .
point in level @xmath23 is recurrent this implies that there must be orbits from @xmath58 to @xmath23 as well .
thus it suffices to show that either ( 1 ) or ( 2 ) holds .
suppose that neither ( 1 ) nor ( 2 ) holds .
consider the set @xmath75 of all points whose code starts with @xmath76 and @xmath77 whose code starts with @xmath78 .
let @xmath79 be the union of the levels @xmath80 to @xmath66 .
since @xmath49 is simple the previous lemma implies that the backwards orbit of every point in @xmath79 must reach @xmath77 , @xmath75 or be singular .
thus we can partition @xmath79 into intervals whose backwards orbit hit @xmath75 , intervals whose backwards orbit hits level @xmath77 , and the backwards singular orbits which divide these intervals .
since we have assumed there are no orbits connecting levels @xmath23 and @xmath58 , corollary [ cor1 ] implies that if the backwards orbit hits @xmath75 ( resp .
@xmath77 ) then the forward orbit hits level @xmath23 ( resp . level @xmath58 ) .
this implies that the points which divide the partition of @xmath79 are also forward singular . since they are both backwards and forward singular they are part of a generalized diagonal , which contradicts that fact that @xmath49 is simple .
therefore at least one of ( 1 ) or ( 2 ) holds , and thus both hold .
next we prove the following strengthening of lemma [ lemma5 ] for simple directions .
this is the first place that the role of the center of the rhombus becomes apparent .
[ lemma5 ] suppose @xmath49 is simple
. then 1 .
there is exactly one code for which @xmath81 and @xmath82 and 2 . for all codes
for which @xmath83 the code is a palindrome and @xmath15 is even .
consider the set @xmath51 of all points whose code starts with @xmath33 and the set @xmath77 of all points whose code starts with @xmath78 .
set @xmath84 .
consider the points in @xmath85 whose orbits arrive at a vertex in @xmath86 before returning to level @xmath32 or @xmath58 .
the cardinality of @xmath86 is @xmath66 , thus since @xmath49 is simple we can apply lemma [ lemma3 ] to conclude that there are exactly @xmath66 such points .
they partition @xmath85 into @xmath87 intervals .
call the associated beams @xmath88 where @xmath89 .
suppose the @xmath63th beam is of length @xmath90 , we denote its code by @xmath91 .
consider the set @xmath92 where @xmath93 is the center of the @xmath63th rhombus .
consider the orbits which start in @xmath85 and arrive at one of these centers before returning to level @xmath32 or @xmath58 . by simplicity
there are @xmath66 such orbits and they are disjoint from the orbits which start in @xmath85 and arrive at a vertex in @xmath86 .
the orbit of @xmath94 is symmetric , thus the code of a beam @xmath88 containing such a point is a palindrome and @xmath15 is even .
the code begins and ends with the same symbol , @xmath32 or @xmath58 , and is strictly positive in between , therefore it can not be a palindrome twice .
thus each of the orbits which arrive at a center are in distinct beams .
this implies that each interval contains at most one @xmath93 and all but two of the beams contain a center .
we have shown that there are two intervals whose beam does not contain a center .
all the other beams start and end on one of the two levels @xmath95 without visiting the other . thus by lemma [ lemma4 ] , the two beams must be as follows : one connects levels @xmath32 to @xmath58 and the other levels @xmath58 to @xmath32 and all other codes are palindromes
. we would like to extend proposition [ lemma5 ] to non simple directions , especially to perpendicular directions .
there are three places where simplicity was used in the proof ,
the first is that the pull back of all @xmath96 and all @xmath94 reaches the set @xmath85 .
a finer counting argument overcomes this difference .
the second is that the pull backs of the @xmath64 and @xmath93 are disjoint .
finally we applied lemma [ lemma4 ] .
we are able to avoid the last two difficulties by assuming that the angle @xmath10 satisfies @xmath97 .
[ babypascal ] fix an irrational right triangle @xmath98 whose smaller angle @xmath97 .
consider the angular coding with respect to the perpendicular direction .
then the first return map to level 0 can not be an iet .
let @xmath23 be the invariant surface containing the perpendicular direction .
consider the set @xmath99 of generalized diagonals which are contained in the invariant surface @xmath23 .
note that since @xmath98 is irrational the set @xmath99 can be infinite .
each generalized diagonal is a segment in @xmath23 .
several generalized diagonals can intersect at a singular point , thus the set @xmath99 has the structure of a graph , not necessarily connected .
suppose the conclusion is not true .
then , the set of perpendicular periodic orbits form a finite union of annuli @xmath100 .
we have @xmath101 since the area of @xmath23 is infinite and the area of @xmath102 is finite .
thus @xmath102 must have a boundary which must consist of a finite union of generalized diagonals .
we will analyze @xmath99 to show that this boundary must be empty .
the boundary of @xmath102 consists of periodic loops .
a _ periodic loop _ is a finite union of generalized diagonals which is periodic .
we can think of a periodic loop as a ( one - sided ) boundary of a family of periodic orbits .
the idea of the proof is to analyze all possible periodic loops which could possibly form the boundary of @xmath102 , i.e. those that have perpendicular periodic orbits on at least one side .
our analysis will show that in each case they must have perpendicular periodic orbits on both sides and thus they can not form part of the boundary of @xmath102 . a periodic loop @xmath103 is called _ simple _
if it consists of a single generalized diagonal , or equivalently the connected component of @xmath99 containing @xmath103 is @xmath104 .
let @xmath105 be the horizontal diagonal of the rhombus .
consider the orbit @xmath106 starting from the left endpoint of @xmath105 .
since @xmath107 the symbolic coding of @xmath106 is @xmath108 , and @xmath106 @xmath106 hits @xmath105 again in an interior point without hitting any other vertex before ( see figure [ fig5 ] ) . thus @xmath106 is a simple periodic loop .
since @xmath106 hits @xmath105 at an interior point it has perpendicular periodic orbits on both sides of it and can not form part of the boundary of @xmath102 .
similarly the perpendicular diagonal starting at the right end point of @xmath105 can not be part of the boundary of @xmath102 .
consider any connected component @xmath109 of @xmath99 such that @xmath110 .
this implies that there is at least one generalized diagonal @xmath111 whose orbit hits @xmath105 perpendicularly , and thus we have a perpendicular periodic loop @xmath112 in @xmath109 . since we have finished treating it , we can assume @xmath113 and thus @xmath114 hits @xmath105 in an interior point .
i claim that the periodic loop hits @xmath105 again perpendicularly at another interior point of @xmath105 .
first of all @xmath115 can not hit @xmath105 again at an endpoint by assumption .
start flowing @xmath116 forwards and backwards from @xmath105 , this is a symmetric orbit in @xmath23 . at exactly half of @xmath115 s ( minimal ) period ,
the forward and backwards orbits meet .
the meeting point must also be a symmetry point of the orbit .
the only points of symmetry available are perpendicular reflections ( in our case since @xmath98 is irrational the only axis of reflection in @xmath23 is @xmath105 ) and central symmetries through copies of the center of @xmath98 .
i claim that the only possibility for the second symmetry of @xmath103 is that it must hit @xmath105 a second time at an different interior point of @xmath105 . first it can not hit a central symmetry point since such points are isolated , while the second symmetry holds for the whole periodic family which @xmath115 bounds .
furthermore @xmath115 can not hit @xmath105 a second time at the same end point , in fact assuming this holds halves the period of @xmath115 , a contradiction of the definition of minimal period .
thus both sides of @xmath114 must consist of periodic perpendicular orbits ( see figure [ fig6 ] ) .
if @xmath115 is simple then both sides of @xmath115 consist of periodic perpendicular orbits and @xmath115 can not be part of the boundary of @xmath102 .
if @xmath115 is not simple then the `` other '' side of @xmath115 is also a periodic loop , for which , by the same reasoning , bounds perpendicular period orbits .
since @xmath102 is a finite union of annuli , there can only be a finite number of perpendicular periodic loops .
thus , by exhaustion , all of @xmath109 consists of perpendicular periodic loops , and can not form part of the boundary of @xmath102 .
this contradiction completes the proof .
next we prove the analog of proposition [ lemma5 ] for perpendicular directions .
[ lemma12 ] fix an irrational right triangle @xmath98 whose smaller angle @xmath97 . consider the angular coding with respect to the perpendicular direction , then 1 .
there is exactly one code for which @xmath54 and @xmath82 and 2 . for all codes
for which @xmath83 the code is a palindrome and @xmath15 is even .
we consider the same notation as in the proof of proposition [ lemma5 ] , the difference being that generalized diagonals exist .
suppose for the moment that no generalized diagonal passes through both @xmath117 and @xmath86 .
note that by symmetry a perpendicular orbit can reach only a single @xmath118 .
let @xmath119 be the cardinality of the points in @xmath85 whose orbits arrive at a vertex in @xmath86 before returning to level @xmath32 or @xmath58 .
they partition the @xmath85 into @xmath120 intervals and corresponding beams @xmath88 .
let @xmath121 be the cardinality of the set of orbits which start in @xmath85 and arrive @xmath117 .
consider the beam @xmath88 containing such an orbit , arguing as in the simple case that each center is in a different beam yields @xmath122 .
if no orbit from @xmath85 reaches @xmath93 before returning level @xmath32 or @xmath58 , then the orbit of @xmath93 never reaches rhombus @xmath32 or @xmath58 .
thus by lemma [ lemma3 ] it s orbit is periodic or its orbit closure is a minimal iet . in either case since the map is a local isometry we can associate @xmath93 with a vertex @xmath123 such that no orbit from @xmath85 reaches this vertex .
furthermore one generalized diagonal can not be the complete boundary to two minimal or periodic components of an iet .
thus for a set of cardinality @xmath105 of such @xmath93 we must have at least @xmath105 distinct such vertices @xmath124 .
this implies @xmath125 , or @xmath126 .
we have shown that @xmath127 , i.e. the number of beams which do not contain a center is at most 2 .
these two beams are the only candidates for beams connecting level @xmath32 to @xmath58 and @xmath58 to @xmath32 .
now we deal with the case when there is an orbit passing through both @xmath117 and @xmath86 .
note that any orbit which starts perpendicularly and arrives at @xmath128 before @xmath86 , returns , by the central symmetry to a perpendicular collision and is thus simple .
thus we have reduced to the case when an orbit visits @xmath86 before @xmath117 . by the central symmetry
it is a generalized diagonal .
the counting argument above works with the following differences .
first of all we consider this generalized diagonal as a degenerate beam @xmath88 . for this degenerate beam
we interpret @xmath44 and @xmath121 as follows . since @xmath88 visits a single @xmath118 it contributes one to @xmath121 and since @xmath120 is the number of beams it contributes one to @xmath44 . with these modifications
the inequality @xmath127 remains true and as above we conclude that all but at most two of the beams contain a center . by the central symmetry around the center the code of each such beam
must satisfy @xmath129 .
next we need to show that exactly two exceptional beams exist . if no exceptional beam exists then lemma [ babypascal ] yields an immediate contradiction . if one exceptional beam exists , if @xmath130 then again lemma [ babypascal ] yields an immediate contradiction . in the other cases ( @xmath131 or @xmath132 )
the points of the exceptional beam can not be recurrent , which is a contradiction of the fact that almost every point is recurrent .
thus exactly two exceptional beams exist .
all the other beams contain a center and thus are palindromes with @xmath15 even .
finally we need to show that in fact one of the two beams satisfies @xmath133 and the other satisfies @xmath134 .
if we think of the beams as marked ( 1st and 2nd ) then there are 16 possible cases depending on of @xmath135 or @xmath58 for @xmath136 and @xmath137 .
two of these cases are good , we must eliminate all the 14 other cases .
since almost every point is recurrent we can not have a beam starting with @xmath32 and ending with @xmath58 without the other beam connecting level @xmath58 to level @xmath32 , this eliminates 6 cases . to be able to apply lemma [ babypascal ] repeat the construction for points which are perpendicular to the leg of the triangle but
whose code is strictly negative before returning to level @xmath32 or @xmath138 to produce two addition exceptional beams with codes @xmath139 , which are centrally symmetric to the two original ones in the sense that @xmath140 for @xmath136 .
a direct application of lemma [ babypascal ] eliminates the the remaining 8 bad cases when there is no beam starting in @xmath32 ending in @xmath58 .
theorem [ thm1 ] the theorem follows immediately for by combining lemma [ lemma2 ] and propostion [ lemma12 ] .
theorem [ thm2 ] we start with the simple case .
first consider orbits whose code starts with @xmath33 .
any such nonsingular nonrecurrent orbit must be contained in the intersection @xmath141 where @xmath142 is the unique interval which satisfy alternative ( 1 ) for fixed @xmath143 of proposition [ lemma5 ] .
consider the initial segment of these beams , these intervals are clearly nested , thus there is at most one interval or a single point in the intersection of their closures .
since the set of nonrecurrent orbits has measure zero it consists of at most a single point .
the orbits of this point is either a positive escape orbits or singular .
similarly we find a single positive escape orbit starting on any level @xmath58 .
this escape orbit must , by uniqueness , coincide with the one starting on level @xmath32 .
a symmetric argument holds for the backwards escape orbit , by starting with points whose code commences with @xmath144 .
we turn to the perpendicular case .
the fist difference to the simple case is that we replace proposition [ lemma5 ] by proposition [ lemma12 ] .
we immediately obtain that there is at most one positive escape orbit starting on level @xmath32 . to show that this is the unique positive escape orbit on the whole surface we argue like in the proof of propostion [ lemma12 ] .
fix @xmath145 .
the first part proof goes through for levels @xmath23 and @xmath58 , i.e. we know that there are at most two exceptional beams .
we need to conclude that they exist and that they connect levels @xmath23 and @xmath58 to each other without applying lemma [ babypascal ] .
this follows immediately from the fact that there is a beam connecting level @xmath32 to @xmath58 and a beam connecting level @xmath58 to level @xmath32 . by symmetry
, the positive and negative escape orbits must be the forward and backwards part of the same orbit .
theorem [ pascal ] the proof is similar to the proof of lemma [ babypascal ] .
the set of perpendicular periodic orbits form a countable union of annuli @xmath100 .
if @xmath101 then @xmath102 must have a boundary which must consist of generalized diagonals and escape orbits .
besides the escape orbit , there is an additional difficulty that the relevant part of the graph of generalized diagonals on @xmath23 need not be finite .
the argument that a _ finite _ connected component @xmath109 satisfying @xmath146 can not form part of the boundary of @xmath102 is the same as in the proof of lemma [ babypascal ] .
this completes the proof in the rational case .
next suppose that the escape orbit exists , i.e. is nonsingular .
denote this orbit by @xmath147 . by construction ,
the orbit @xmath147 starts on level @xmath32 at a point @xmath148 which is an interior point of @xmath105 .
thus we can find a sequence of perpendicular periodic orbits which approaches @xmath147 from the left and a sequence which approaches @xmath147 from the right , and @xmath147 can not form part of the boundary of @xmath102 . the remaining case is @xmath109 infinite .
there are two subcases , either every edge in the graph has periodic perpendicular orbits on both sides or not . in the first case we argue as in the proof of lemma [ babypascal ] , using induction instead of exhaustion to complete the proof . in the second case the ( unique )
escape orbit @xmath147 is infact singular and contained in @xmath109 , see figure [ fig7 ] .
thus the one side of @xmath147 bounds perpendicular periodic families , while the other side is a limit or perpendicular periodic families .
the method of proof also allows to verify the boshernitzan conjecture for an infinite collection of rational right triangles , those where the perpendicular orbits starting at the ( non - right angle ) verteces are simple periodic loops .
theorem [ thm0 ] for irrational triangles this follows immediately from theorem [ pascal ] since the surface @xmath23 is dense in the full phase space . for rational triangles the result was established in @xcite .
i would like to remark on the assumption @xmath11 .
this is a technical assumption which guarantees that the orbits starting at the endpoints of @xmath105 are simple periodic loops .
this is the key condition which is used in the proof of the theorems .
it seems likely this condition holds for an open set of full measure of parameters @xmath10 , but to prove this seems messy and complicated .
i profitted greatly from discussions with pascal hubert . | there is an open set of right triangles such that for each irrational triangle in this set ( i ) periodic billiards orbits are dense in the phase space , ( ii ) there is a unique nonsingular perpendicular billiard orbit which is not periodic , and ( iii ) the perpendicular periodic orbits fill the corresponding invariant surface .
[ 1]*proof of # 1 .
* + |
supernova type 1a ( snia ) distance modulus measurements have grown into a powerful measurement of the equation of state of our universe .
accurate cosmological distances combined with redshift measurements allow for a precise characterisation of the hubble flow as well as the additional acceleration attributed to the cosmological constant @xcite .
the increasing statistics of snia measurements together with more information for each separate supernova event have progressively lowered the observational uncertainties : dust extinction , photometric error and lightcurve characterisation ( see e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
extinction by dust remains a problematic systematic uncertainty in snia observations because the applicable extinction law remains poorly understood .
this will need to be addressed in order to use snia in the next step in accuracy for a cosmological probe .
dust attenuation affects both the observed snia rate and the distance modulus .
extinction by dust occurs in three instances before observation of the snia light : ( 1 ) in our own milky way , ( 2 ) in intergalactic space , and ( 3 ) in the host galaxy of the snia . galactic extinction ( 1 ) is a well - studied problem , because it is an ubiquitous one .
@xcite produced a map of galactic extinction based on hi maps and distant galaxy counts .
it was superseded by the map of @xcite , based on cobe and iras far - infrared and sub - mm maps .
the latter map is now generally used , together with the galactic extinction law to correct extragalactic sources @xcite ; the inferred extinction along a line - of - sight is proportional to the reddening : @xmath4 .
the canonical value of @xmath0 is 3.1 with occasionally lower values towards the galactic center @xcite and higher elsewhere @xcite .
extinction by dust in intergalactic space ( 2 ) has been proposed as an alternative explanation for snia dimming , which is generally attributed to the cosmological acceleration @xcite .
the resulting extinction law of this dust would effectively be grey because the attenuating dust would be spread over all redshifts along every line - of - sight .
the coincidence of both uniform distribution in both sky and redshift space does give this explanation a somewhat contrived appearance .
the injection of dust into the intergalactic medium would have to be constant and substantial .
models of a dusty universe @xcite find this grey dust explanation increasingly inconsistent with observational data ( see also e.g. , * ? ? ?
extinction within the snia s host galaxy ( 3 ) , dust in the immediate surroundings and any disk , ring or spiral arm the line - of - sight passes through in projection , is an observationally evident , yet not fully constrained uncertainty .
the dark energy task group report @xcite notes this as a primary source of uncertainty for snia measurements .
three characteristics of the host galaxy s dust could and are expected to change over the history of the universe : ( a ) total dust content or mass , ( b ) dust distribution within the host galaxy and ( c ) dust composition .
the overall effect on the effective extinction law is of interest for the distance determination from snia light curves .
dust mass ( a ) is a variable in several spectral energy distribution ( sed ) studies of distant galaxies .
@xcite modeled the overall dust content of galaxies over time and found that a maximum occurred either at z=1 or at z=3 .
@xcite modeled the sed s from distant galaxies and similarly found a maximum dust content at z=1 .
@xcite find a steady increase of dust mass with time from the uv - ir sed of galaxies .
the typical dust mass found in distant galaxies is very much a function of the selected sample .
far - infrared selected samples point to dust - rich galaxies , similar to arp 220 @xcite , optical / uv selected samples point to disks very similar to the local ones @xcite and lyman-@xmath5 galaxies point to low - extinction disks @xcite .
however , more dust mass should not affect the snia distance determinations if the extinction law remains the same for nearby and distant galaxies .
more dust in the distant galaxies will predominantly affect the observed snia _ rate _
@xcite , as heavily obscured snia drop from observed samples .
the dust distribution in host galaxies ( b ) is commonly modeled as a double exponential , one radial and one vertical , sometimes with rings to mimic spiral arms .
the radial scale of the dust distribution is assumed to be similar to the stellar one and the vertical dust scale is supposed to be much smaller than the stellar one .
previous observations of the scale - height in nearby ( edge - on ) galaxies appeared to corroborate the small scale - height ( e.g , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) but recent observations indicate a much higher scale - height for the dust @xcite , similar to the stellar scale .
if the average scale - height of the dust distribution was higher in the past , then snia in the plane of the disk will encounter more extinction , especially when viewed in an inclined system .
the different distribution has a similar effect as variation the dust content of galaxies and only the observed snia rate will be affected , unless dust composition and distribution are related . the dust composition ( c ) , notably the ratio of small to large grains , directly affects the observed extinction law ( see the review by * ? ? ?
evolution in the average extinction law is the most troubling possibility , as this would affect the extinction correction of snia and indirectly the measured hubble flow and acceleration . for local snia , variations in the extinction law
have been observed ( e.g. , * ? ? ?
* ; * ? ? ?
@xcite explained the different observed extinction law for some snia as the effect of circumstellar material around the snia progenitor .
this is a plausible scenario as the massive - star supernovae are very efficient dust producers @xcite .
@xcite and @xcite report observations of such material .
alternatively , there is substantial evidence for a link between star - formation and extinction law in star - forming galaxies ( see the review in * ? ? ? * and the references therein . ) .
star - formation produces more small grains and the intense uv fields alter grain composition . since it is not unreasonable to suppose that star - formation affects both the distribution and the extinction characteristic of dust in spiral disks , i propose a simple model for the evolution of the extinction law applicable to snia measurements .
the aim of this paper is to explore this link between the extinction law for snia and star - formation of host galaxies and to investigate how much extinction law evolution could reasonably influence snia distance measurements .
the paper is organized in a brief review of current observations of evolution in host galaxy dust and @xmath0 ( 2 ) , a description of the model ( 3 ) , two tests based on current snia observations ( 4 ) , and discussion and conclusions ( 5 ) .
observational evidence for a different extinction law at higher redshifts comes from lensed quasars , qso reddening by damped lyman-@xmath5 systems and grb afterglows , as well as snia measurements .
gravitational lenses of qsos find a range of values for @xmath0 up to a redshift of 1 @xcite .
however , the lenses are often elliptical galaxies , which are not typical snia hosts .
@xcite find evidence of qso reddening by lyman-@xmath5 systems but this is disputed by @xcite , both based on the sdss sample .
@xcite find a smc - type extinction based on all sdss qsos with a lyman-@xmath5 system .
@xcite compare extinciton for different grb afterglows and find evidence for dust in the grb hosts .
none of these observations effectively constrain the extinction law applicable to snia . in the snia literature , the problem of snia host galaxy dust extinction is well recognised @xcite .
observations of high redshift snia tell us that their spectra are similar to local ones @xcite , with no anomalous reddening @xcite and dust masses of host galaxies similar to those of local galaxies ( * ? ? ?
* ; * ? ? ?
* based on sub - mm data ) .
several studies look at the relations between snia lightcurve properties ( peak brightness , duration , and colour ) and host galaxy properties ( type , stellar mass , extinction , and star - formation ) . a relation between sn peak brightness and host galaxy type
is well established @xcite .
@xcite refine the relation to one between the snia lightcurve and star - formation in the galaxy : passive galaxies , with no star - formation , preferentially host faster - declining / dimmer sn ia , while brighter events are found in systems with ongoing star - formation .
the snia related to ongoing star - formation could well dominate the population at higher redshift @xcite , as there is much more star - formation .
host galaxy extinction is of equal importance as any change in the snia population .
@xcite report dimmer snia for spiral galaxy hosts , especially when projected in or through the inner part of the disk .
@xcite also finds a radial dependence of the extinction of local snia .
values for @xmath0 different than the canonical @xmath0 = 3.1 have been found for individual snia @xcite .
based on their entire sample , @xcite find a value of 2.65 .
@xcite find a mean value of 2.7 for heavily obscure sn and 2.9 for the less extincted ones .
the explanations for deviant @xmath0 values are circumstellar material @xcite or a different dust composition @xcite .
evolution in @xmath0 may be critical for precision cosmology , and the snia themselves are also the best probe to characterize this evolution .
values out of the disk of the host galaxy . a line - of - sight to a snia
can intersect such plumes .
the higher star - formation at z@xmath61 results in more plumes and hence a higher probability for a snia line - of - sight intersecting one .
, scaledwidth=45.0% ]
let us consider four observational facts : ( i ) snia occur predominantly in spirals @xcite , ( ii ) star - formation ejects ism from the planes of spiral disks and the ejected material takes some time to rain back @xcite , ( iii ) star - formation modifies the extinction characteristics ( lower @xmath0 , * ? ? ?
* ; * ? ? ?
* ) and ( iv ) there is an order of magnitude more star - formation at and beyond z=1 @xcite .
therefore , at z@xmath61 , spiral disks are likely to have an halo of ism , recently processed and ejected by star - formation . for the distance measurement with snia this is especially important as their light will encounter more @xmath7 type dust as a result ( see figure [ f : cartoon ] ) .
hence , the _ average _ value of @xmath0 ( @xmath1 ) might be lower for snia at higher redshift .
the simplest parametrisation of the @xmath1 evolution is a second order polynomial : @xmath8 with the parameters depending on the choice of @xmath1 value for the different epochs .
of the four assertions above , the relation between star - formation and extinction law ( iii ) is the most tenuous .
evidence for a different extinction law in star - forming galaxies have been found by @xcite .
however , for extended sources such as hii regions , one would need to disentangle the effects of processed dust grains and a clumpy medium ( see * ? ? ?
@xcite presents extinction law measurements for single stars , and hence lines - of - sight , in different regions in the smc and lmc .
they find evidence of dust processing by star - formation and lower values of @xmath1 in the star - forming parts .
their average @xmath0 values are : smc bar ; @xmath9 , lmc supershell ; @xmath10 , lmc average sample ; @xmath11 .
i note that the values for the regions with higher star - formation are lower than the average for the lmc . in this paper
i explore three models : model a assumes no evolution in @xmath1 .
model b takes the lmc and smc values as a template for the galaxies at z=1 and earlier : star - forming galaxies at z@xmath61 have @xmath1 = 2.7 and gas - rich galaxies at z@xmath62 and beyond have @xmath12 .
model c leaves the parameters in equation 1 free to fit the distributions of snia extinction values ( @xmath2 ) in a low , intermediate and high redshift sample of snia observations .
there are two suggestive observations of snia that support a general model of evolution in host galaxy extinction : ( a ) the relation between @xmath1 for snia and host galaxy disk inclination and ( b ) the distribution of inferred extinction ( @xmath2 ) for a sample of high , intermediate and low redshift snia .
values from @xcite , for the late - type galaxies .
top panel : the number of snia with @xmath13 and the number with deviant values ( dashed line ) .
middle panel : the percentage of snia with deviant @xmath0 values .
bottom panel : the average value of @xmath0 .
there are very few sn observed in perfectly face - on galaxies . in higher inclined disks the average value of @xmath0 is lower , indicating a possible effect of ejecta in the galaxy s halo.,scaledwidth=45.0% ] @xcite
use the multi - filter lightcurves available for a sample of 133 snia to independently fit the value of @xmath0 as well as the light - curve peak and snia colour .
the inclination and host galaxy type are from _ hyperleda _ and @xmath0 values from table 4 in @xcite .
figure [ f : incl ] shows the distribution of these snia as a function of disk inclination of the late - type host galaxies for those snia with @xmath14 ( the default ) and those with deviant values ( @xmath15 ) .
notably , the deviant values start to dominate at the higher inclinations .
figure [ f : incl ] also shows the @xmath1 value for each inclination bin ; it diminishes with increasing inclination .
this supports our model of recent ejecta lowering the value of @xmath1 .
the cumulative histogram of snia reddening for three samples : the nearby snia sample ( * ? ? ?
* solid line ) , the intermediate distance sample from the essence survey ( * ? ? ?
* dotted line ) and the high redshift sample ( * ? ? ?
* dashed line ) .
the differences in distribution point to evolution of the extinction in host galaxies , either in @xmath2 , @xmath0 or both.,scaledwidth=45.0% ] the second test is to compare the distributions of @xmath2 of snia at different redshift .
figure [ f : ebv ] shows the distribution of extinction values ( @xmath2 ) , based on reddening and a @xmath0 valued of 3.1 , for three samples of snia , a local , an intermediate and a high - redshift one .
local snia are those from the sample of @xcite .
the intermediate redshift sample is the essence data from @xcite and the highest redshift data is from @xcite .
sample sizes are very different : 133 , 189 and 33 for the low , intermediate , and high redshift samples respectively .
the assumption is that the reddening measurements are comparable between these samples .
each of these studies uses a ` prior ' distribution for the extinction values to optimize the fit to the lightcurves and i assume here that the affect of this prior is negligible on the reported distribution ( see for a good discussion of extinction priors * ? ? ?
the distributions of extinction values in figure [ f : ebv ] show that there is some type of evolution in @xmath2 from one redshift sample to the next .
snia searches use a gold " and silver " standard for snia lightcurves . to mimick these
i limit the samples to @xmath16 and @xmath17 respectively in the following analysis .
top panels : the three models considered for evolution of @xmath3 , ( a ) no @xmath1 evolution , ( b ) @xmath18 with @xmath19 and @xmath20 , and ( c ) with @xmath18 fit to minimize the difference between the distribution of @xcite and the other two distributions ( @xmath21 = 0.25 to 0.3 ) = 3.2 to 4.1 ) .
bottom panels : the cumulative histogram of @xmath2 of snia from a nearby ( solid line * ? ? ?
* ) , intermediate ( dotted line * ? ? ?
* ) and distant sample ( dashed line * ? ? ?
* ) for the different models in the top panels ) .
, scaledwidth=45.0% ] in figure [ f : avmod ] , i compare the three different models for @xmath0 evolution ( top panels ) using the recomputed @xmath2 distribution for each model ( bottom panels ) .
i limit the comparison to the snia silver " standard to maximize statistics . figure [ f : avmod ] shows model a ( no evolution , @xmath1 = 3.1 ) and two evolution models ( b and c ) .
model b uses the magellanic cloud values from @xcite to parameterise the dependence of @xmath1(z ) .
model c uses the best fit of equation 1 to minimize the differences between the @xmath2 distributions , assuming for each sample the same parent distribution , i.e. , random lines of sight through a disk of a spiral galaxy .
the kolmogorov - smirnov probability that all three @xmath2 populations are from the same parent populations without a change in @xmath1 ( model a , figure [ f : avmod ] , top left panel ) is extremely low ( @xmath22 , figure [ f : avmod ] , bottom left panel ) .
hence , some type of evolution is present in the distribution of snia extinction values ( @xmath2 ) or extinction law ( @xmath1 ) .
model b is what one naively would expect on the basis of the narrative in 3 : the star - forming galaxies have a lower @xmath1 , which rises with gas - fraction at higher z and using the @xcite values as a template .
the top middle panel in figure [ f : avmod ] shows the @xmath1 values .
i recomputed the values for @xmath2 with the model @xmath3 ( figure [ f : avmod ] , bottom middle panel ) .
this model does not significantly improve the match between the three distributions of snia extinctions at different redshifts ( see table [ t : mod ] ) .
if this model is true , there is substantial independent evolution in the distribution of @xmath2 .
model c is a fit of the parameters in equation 1 to recompute the @xmath2 distribution and maximizing the probability that they are from the same parent distribution .
the best fit values are in table [ t : mod ] and figure [ f : avmod ] shows the inferred @xmath1 ( top right ) and computed @xmath2 distribution ( bottom right ) . the best model to match the intermediate and high - redshift samples to the low - redshift reference peaks at @xmath23 = 0.25 to 0.3 with a high value of @xmath24 .
in this case little or no evolution in @xmath2 is needed . however , the values of @xmath1 for each epoch are counter to expectations of 3 . the maximum possible difference in extinction correction for the silver " ( @xmath2 = 0.5 for all snia ) and gold " ( @xmath2 = 0.25 for all snia ) samples between @xmath14 and model c. the points are the binned snia residuals , after subtracting the empty universe model , from @xcite , figure 6 . even if all the snia are affected by the maximum @xmath2 _ and _
model c holds true , evolution in @xmath1 _ alone _ can not explain the residuals in average distance moduli of snia.,scaledwidth=45.0% ] figure [ f : dm ] shows the observed average dimming of snia as a function of redshift with the maximum possible impact of @xmath1 evolution plotted : model c with the maximum e(b - v ) value for each snia allowed by the silver and gold sample selection . in this worst case scenario
i assume that all snia are affected by the maximum reddening allowed for selection into the gold or silver samples , even though they are clearly not ( figure [ f : ebv ] ) .
the worst - case scenario ( pure @xmath1 evolution ) has the same shape as the dimming observed in snia , but the effect of @xmath1 evolution is not strong enough to fully account for the observed dimming of snia .
the exact origin of @xmath1 evolution can be either changes in the immediate surroundings of snia or evolution in the ism of host galaxies . because high - redshift observations of snia lightcurves are often limited to two filters
to minimize observatory time spent on a given object , the applied extinction correction is mostly the fiducial @xmath14 .
occasional evidence from higher - redshift spectra already point to very low values of @xmath0 ( a. riess , private communication ) .
figure [ f : dm ] shows that multi - filter and/or spectroscopic determination of @xmath0 for observed snia are crucial to accurately determine the equation of state of our universe from snia distances .
it seems likely that the distribution of @xmath2 values for snia lightcurves will vary as a function of redshift . in this paper ,
i treated this as evolution of the average extinction law ( @xmath1 ) .
there , however , are three possible explanations for the behavior of the @xmath2 distribution , apart from a fit - prior artifact : * @xmath1 does not change much , but due to dust ejection and/or additional dust production ( from the elevated star - formation in each galaxy ) , each snia at higher redshift is seen through more host galaxy dust . *
the host galaxies are similar to nearby spirals but snia occur preferentially in different environments at each redshift . at higher redshifts ,
they occur in more dust - rich parts of spirals ( e.g. , the spiral arms or in interacting galaxies ) .
* @xmath1 and hence @xmath2 does evolve as a combination of its dependence on dust composition and dust distribution in the host galaxy .
options 1 and 2 mean that the observed difference in @xmath2 distribution ( figure [ f : avmod ] ) at different redshifts is real and the inferred distances from snia lightcurves suffer no ill effects
. in option 3 , however , the evolution in the host galaxy extinction law could skew the observations of the acceleration of the expansion of the universe and the inferred cosmology .
i would like to stress that option 3 is not as far - fetched as cosmic gray dust and even the worst case scenario would not fully explain the snia dimming ( figure [ f : dm ] )
. however , a realistic model would require ( 1 ) the relative positions in the disk of snia at higher redshifts to characterize the importance of scenario 2 , and ( 2 ) fits of their lightcurves which include @xmath0 as a fit parameter , similar to those of @xcite , to constrain scenario 3 .
given enough statistics , the snia measurements can then be used as probes of the distribution and severity of host galaxy extinction in disks at high redshift . in this paper
i assumed that the distributions of reddening values are uniformly derived and the choice of prior in the lightcurve analysis did not matter . in future work
, one would prefer the same prior in all redshift samples and prefer even more sufficient information to conclusively determine @xmath2 and @xmath0 for each snia lightcurve . in conclusion
: * there is a second plausible mechanism that modifies the extinction law for snia in higher redshift host galaxies ( 3 ) ; host galaxy dust evolution in addition to material around the snia progenitor . * the toy model of this mechanism
does not match the @xmath2 distribution of snia samples at different redshifts ( 4 , figure [ f : avmod ] , model b ) . *
matching up @xmath2 distributions leads to a @xmath1 evolution model that resembles the shape and peak of the observed average dimming of snia ( 4 , figure [ f : avmod ] , model c ) . *
taken as a worst - case scenario , this model still can not account for the observed average dimming of snia with redshift . * it is likely that there is evolution in the distribution of @xmath2 values with redshift , making it impossible to deduce @xmath1 evolution from these distributions alone .
as both affect the precision cosmology to be done with the next generation of snia measurements , it is paramount that new snia observations solve for both @xmath2 and @xmath0 separately . *
the bonus science of measuring both @xmath2 and @xmath0 for each snia is a probe in the distribution , content and composition of dust in their host galaxies .
the author would like to thank michael wood - vasey , andy fruchter , rosa gonzles , daniela calzetti , ron allen and adam riess for useful discussions and comments on snia host galaxy extinction . .[t
: mod]the values for two different toy models , the peak redshift and peak value of @xmath1 and the kolmogorov - smirnov test [ cols= " < , < , < , < , < , < , < " , ] | this paper presents a mechanism that may modify the extinction law for snia observed at higher redshift .
starting from the observations that ( 1 ) snia occur predominantly in spiral galaxies , ( 2 ) star - formation ejects ism out of the plane of spirals , ( 3 ) star - formation alters the extinction properties of the dust in the ism , and ( 4 ) there is substantially more star - formation at higher redshift , i propose that spiral galaxies have a dustier halo in the past than they do now . the ejected material s
lower value of @xmath0 will lead to a lower average value ( @xmath1 ) for snia observed at higher redshift .
two relations in snia observations indicate evolution of the average @xmath0 : the relation of observed @xmath0 with inclination of the host galaxy at low redshift and the matching of the distribution of extinction values ( @xmath2 ) for snia in different redshift intervals .
the inclination effect does point to a halo with lower @xmath0 values .
in contrast , the distributions of @xmath2 values match best for a @xmath3 evolution that mimics the relation of snia dimming with redshift attributed to the cosmological constant .
however , even in the worse case scenario , the evolution @xmath1 can not fully explain the dimming of snia : host galaxy extinction law evolution is not a viable alternative to account for the dimming of snia .
future observations of snia multi - color lightcurves and spectra will solve separately for values of @xmath2 and @xmath0 for each snia .
solving for evolution of @xmath1 ( and @xmath2 ) with redshift will be important for the coming generation of cosmological snia measurements and has the bonus science of insight into the distribution of dust - rich ism in the host galaxies in the distant past .
[ firstpage ] ( cosmology : ) distance scale cosmology : observations galaxies : high - redshift galaxies : ism ( ism : ) dust , extinction ( stars : ) supernovae : general |
adiabatic quantum computation ( aqc ) was proposed by farhi et al.@xcite in 2000 .
the aqc model is based on the _ adiabatic theorem _ ( see , e.g. @xcite ) .
informally , the theorem says that if we take a quantum system whose hamiltonian `` slowly '' changes from @xmath3 ( initial hamiltonian ) to @xmath4 ( final hamiltonian ) , then if we start with the system in the _ groundstate _
( eigenvector corresponding to the lowest eigenvalue ) of @xmath3 , then at the end of the evolution the system will be `` predominantly '' in the ground state of @xmath4 . the theorem is used to construct _
adiabatic algorithms _ for optimization problems in the following way : the initial hamiltonian @xmath3 is designed such that the system can be readily initialized into its known groundstate , while the groundstate of the final hamiltonian @xmath4 encodes the answer to the desired optimization problem .
the complete ( or _ system _ ) hamiltonian at a time @xmath5 is then given by @xmath6 for @xmath7 $ ] where @xmath8 increases monotonically from @xmath9 to @xmath10 and @xmath11 is the total running time of the algorithm .
if @xmath11 is large enough , which is determined by the minimum spectral gap ( the difference between the two lowest energy levels ) of the system hamiltonian , the adiabatic theorem guarantees the state at time @xmath5 will be the groundstate of @xmath12 , leading to the solution , the ground state of @xmath13 .
it is believed that aqc is advantageous over standard ( the gate model ) quantum computation in that it is more robust against environmental noise @xcite . in 2004 , d - wave systems inc . undertook the endeavor to build an adiabatic quantum computer for solving np - hard problems .
kaminsky and lloyd @xcite proposed a scalable architecture for aqc of np - hard problems .
note that aqc is polynomially equivalent to the standard quantum computation , and therefore aqc is universal .
farhi et al . in their original paper
@xcite showed that aqc can be efficiently simulated by standard quantum computers . in 2004 , aharonov et al @xcite proved the more difficult converse statement that standard quantum computation can be efficiently simulated by aqc . in this paper , however , we will focus on a subclass of hamiltonians , known as _ ising models in a transverse field _ , that are np - hard but not universal for quantum computation .
this subclass of hamiltonians has been implemented by d - wave systems inc .
d - wave s quantum hardware architecture can be viewed as an undirected graph @xmath1 with weighted vertices and weighted edges .
see figure [ fig : extendedgrid ] for an example .
denote the vertex set of @xmath1 by @xmath14 and the edge set of @xmath1 by @xmath15 .
each vertex @xmath16 corresponds to a qubit , and each edge @xmath17 corresponds to a coupler between qubit @xmath18 and qubit @xmath19 . in the following
, we will use qubit and vertex , and coupler and edge interchangeably when there is no confusion .
there are two weights , @xmath20 ( called the qubit _ bias _ ) and @xmath21 ( called the _ tunneling amplitude _ ) , associated with each qubit @xmath18 .
there is a weight @xmath22 ( called the coupler _ strength _ ) associated with each coupler @xmath23 . in general
, these weights are functions of time , i.e. , they vary over the time , e.g. , @xmath24 . the hamiltonian of the ising model in a transverse field thus implemented is : @xmath25 with @xmath26 ( the @xmath27 is in the @xmath18th position ) , similarly for @xmath28 and @xmath29 , where @xmath30 is the @xmath31 identity matrix , while @xmath27 and @xmath32 are the pauli matrices given by @xmath33 in general , the transverse field ( i.e. @xmath34 ) encodes the initial hamiltonian @xmath3 , while the other two terms encode the final hamiltonian : @xmath35 one can show ( see appendix ) that the eigenvalues and the corresponding eigenstates of @xmath4 are encoded in the following energy function of ising model : @xmath36 where @xmath37 , called a _ spin _ , and @xmath38 .
in particular , the smallest eigenvalue of @xmath4 corresponding to the minimum of @xmath39 , and @xmath40 corresponds to its eigenvector ( called _ ground state _ ) of @xmath4 .
when there is no confusion , we use the energy function of ising model and ising hamiltonian interchangeably .
hereafter , we refer the problem of finding the minimum energy of the ising model or equivalently the groundstate of ising hamiltonian as the _ ising problem_. it can be shown that the ising problem is equivalent to the problem of quadratic unconstrained boolean optimization ( qubo)(see @xcite and references therein ) , which has been shown to be a common model for a wide variety of discrete optimization problems .
more specifically , finding the minimum of @xmath39 in is equivalent to finding the maximum of the following function ( which is also known as quadratic pseudo - boolean function @xcite ) of qubo on the same graph : @xmath41 where @xmath42 , @xmath43 .
the correspondence between the parameters , @xmath44s and @xmath45s will be shown in section [ sec : qubo - ising ] .
therefore , given an ising / qubo problem on graph @xmath0 , one can thus solve the problem on an adiabatic quantum computer ( using an ising spin-1/2 system ) if @xmath0 can be embedded as a subgraph of the quantum hardware graph @xmath1 .
we refer this embedding problem as _ subgraph - embedding _ , to be defined formally in section [ sec : embed ] .
in general , there are physical constraints on the hardware graph @xmath1 . in particular
, there is a _ degree - constraint _ in that each qubit can have at most a constant number of couplers dictated by hardware design .
therefore , besides the possible difficulty of the subgraph - embedding problem , the graphs that can be solved on a given hardware graph @xmath1 through subgraph - embedding must also be degree - bounded .
kaminsky et al .
@xcite observed and proposed that one can embed @xmath0 in @xmath46 through ferromagnetic coupling `` dummy vertices '' to solve maximum independent set ( mis ) problem .
] of planar cubic graphs ( regular graphs of degree-3 ) on an adiabatic quantum computer . in particular
, they proposed an @xmath47 square lattice as a scalable hardware architecture on which all @xmath48-vertex planar cubic graphs are embeddable .
the notion of embedding here follows naturally from physicists intuition that each _ logical qubit _ ( corresponding to a vertex in the input graph ) is mapped to a subtree of _ physical qubits _ ( corresponding to vertices in the hardware graph ) that are ferromagnetically coupled such that each subtree of physical qubits acts like a single logical qubit .
for example , in figure [ fig : example - embedding ] , the logical qubit @xmath49 ( in orange color ) of the graph @xmath0 is mapped to a subtree of physical qubits ( labelled @xmath49 ) of the square lattice . informally , a minor - embedding @xmath2 of a graph @xmath0 in the hardware graph @xmath1 is a subgraph of @xmath1 such that @xmath2 is an `` expansion '' of @xmath0 by replacing each vertex of @xmath0 with a ( connected ) subtree of @xmath1 , or equivalently , @xmath0 can be obtained from @xmath2 by contracting edges ( same color in figure [ fig : example - embedding ] ) . in graph theory
, @xmath0 is called a ( graph ) _ minor _ of @xmath1 ( see for example @xcite ) .
the minor - embedding will be formally defined in section [ sec : embed ] .
( remark : the embedding in @xcite is a special case of minor - embedding , known as _ topological - minor _ embedding . ) by _ reduction through minor - embedding _ , we mean that one can reduce the original ising hamiltonian on the input graph @xmath0 to the embedded ising hamiltonian @xmath50 on its minor - embedding @xmath2 , i.e. , the solution to the embedded ising hamiltonian gives rise to the solution to the original ising hamiltonian .
the intuition suggests that the reduction will be correct provided that the ferromagnetic coupler strengths used are sufficiently strong ( i.e. , large negative number ) .
however , how strong is `` strong enough '' ? in
@xcite , they do not address this question ,
i.e. what are the required strengths of these ferromagnetic couplers ? in section [ sec : easy - upper ] , we will show that it is not difficult to give an upper bound for the ferromagnetic coupler strengths and thus explain the intuition .
however , there are indications @xcite that too strong ferromagnetic coupler strengths might slow down the adiabatic algorithm .
furthermore , an adiabatic quantum computer is an analog computer and analog parameters can only be set to a certain degree of precision ( a condition much more stringent than the setting of digital parameters ) .
hence , the allowed values of coupler strengths are limited .
therefore , from the computational point of view , it is important to derive as small ( in terms of magnitude ) as possible sufficient condition for these ferromagnetic coupler strengths .
furthermore , what should the bias for physical qubits be ?
there are two components to the reduction : _ embedding _ and _ parameter setting_. the embedding problem is to find a minor - embedding @xmath2 of a graph @xmath0 in @xmath1 .
this problem is interdependent of the hardware graph design problem , which will be discussed in section [ sec : discussion ] .
the parameter setting problem is to set the corresponding parameters , qubit bias and coupler strengths , of the embedded ising hamiltonian . in this paper
, we assume that the minor - embedding @xmath2 is given , and focus on the parameter setting problem ( of the final hamiltonian ) .
note that there are two aspects of efficiency of a reduction .
one is how efficient one can reduce the original problem to the reduced problem .
for example , here we are concerned how efficiently we can compute the minor - embedding and how efficiently we can compute the new parameters of the embedded ising hamiltonian .
the other aspect concerns about the efficiency ( in terms the running time ) of the adiabatic algorithm for the reduced problem . in general
, the latter depends on the former .
for example , the running time of the adiabatic algorithm may depend on a `` good '' embedding that is reduced to . according to the adiabatic theorem ( see , e.g. @xcite ) ,
the running time of the adiabatic algorithm depends on the minimum spectral gap ( the difference between the two lowest energy levels ) of the system hamiltonian , which is defined by both initial hamiltonian and final hamiltonian .
that is , the running time ( and thus the efficiency of the reduction ) will depend on both the initial hamiltonian and the final hamiltonian . in this paper ,
our focus is only on the final hamiltonian , and therefore we are not able to address the running time of the adiabatic algorithm .
furthermore , the estimation of the minimum spectral gap of the ( system ) hamiltonian is in general hard .
consequently , analytically analyzing the running time of an adiabatic algorithm is in general an open question .
finally , let us remark that there is another different approach based on perturbation theory by oliveira & terhal @xcite for performing the reduction .
in particular , they employed perturbative gadgets to reduce a 2-local ( system ) hamiltonian to a 2-local hamiltonian on a 2-d square lattice , and were able to show ( as in the pioneering work @xcite ) that the minimum spectral gap ( and thus the running time ) of the system hamiltonian is preserved ( up to a polynomial factor ) after the reduction , for any given initial hamiltonian .
however , besides the ineffective embedding , as pointed out in @xcite , the method is `` unphysical '' as it requires that each parameter grows with the system size .
the rest of the paper is organized as follows . in section [ sec : qubo - ising ] , we recall the equivalences between the qubo problem and the ising problem . in section
[ sec : embed ] , we introduce the minor - embedding definition and mention related work in graph theory . in section [ sec : parameters ] , we derive the new parameters , namely values for the qubit bias and the sufficient condition for the ferromagnetic coupler strengths , for the embedded ising hamiltonian such that the original ising problem can be correctly solved through the embedded ising problem . in section [ sec : wmis ] , we show the embedded ising hamiltonian for solving the ( weighted ) mis problem . finally , we conclude with several related algorithmic problems that need to be investigated in order to facilitate the design of efficient adiabatic algorithms and aqc architectures in section [ sec : discussion ] .
in this section , we recall the equivalences between the problem of qubo ( maximization of @xmath51 in eq . ) and the ising problem ( minimization of @xmath39 in eq . ) .
notice that @xmath52 , that is , @xmath53 corresponds to @xmath54 , and @xmath55 corresponds to @xmath56 . using a change of variables
, we have @xmath57 where @xmath58 , the neighborhood of vertex @xmath18 , for @xmath59
. therefore , @xmath60 in eq . is equivalent to @xmath61 in eq .
where @xmath62 , or @xmath63 for @xmath59 .
see figure [ fig : new - parameters ] for the correspondences between parameters in qubo and the ising model .
let @xmath1 be a fixed hardware graph .
given @xmath0 , the _ minor - embedding _ of @xmath0 is defined by @xmath64 such that * each vertex in @xmath65 is mapped to a connected subtree @xmath66 of @xmath1 ; * there exists a map @xmath67 such that for each @xmath68 , there are corresponding @xmath69 and @xmath70 with @xmath71 .
given @xmath0 , if @xmath72 exists , we say that @xmath0 is _ embeddable _ in @xmath1 . in graph theory , @xmath0 is called a _
minor _ of @xmath1 .
when @xmath72 is clear from the context , we denote the minor - embedding @xmath73 of @xmath0 by @xmath2 .
equivalently , one can think of a minor @xmath0 of @xmath1 as a graph that can be obtained from a subgraph of @xmath1 by contracting edges .
see figure [ fig : example - embedding ] for an example .
in particular , there are two special cases of minor - embedding : * _ subgraph - embedding : _ each @xmath66 consists of a single vertex in @xmath1 .
that is , @xmath0 is isomorphic to @xmath2 ( a subgraph of @xmath1 ) . * _ topological - minor - embedding : _ each @xmath66 is a chain ( or path ) of vertices in @xmath1 .
minors are well - studied in graph theory , see for example @xcite .
given a _ fixed _
graph @xmath0 , there are algorithms that find a minor - embedding of @xmath0 in @xmath1 in polynomial time of size of @xmath1 , from the pioneering @xmath74 time algorithm by robertson and seymour @xcite to recent nearly linear time algorithm of b. reed ( not yet published ) .
however , it is worthwhile to reiterate that these algorithms are for _ fixed _ @xmath0 , and their running times are _ exponential _ in the size of @xmath0 . here
the minor - embedding problem is to find a minor - embedding of @xmath0 ( for any given @xmath0 ) while fixing @xmath1 .
to the best of our knowledge , the only known work related to our minor - embedding problem was by kleinberg and rubinfeld @xcite , in which they showed that there is a randomized polynomial algorithm , based on a random walk , to find a minor - embedding in a given degree - bounded expander .
our embedding problem might appear similar to the embedding problem from parallel architecture studies .
however , besides the different physical constraints for the design of architectures , the requirements are very different . in particular , in our embedding problem
, we do not allow for _ load _ @xmath75 , which is the maximum number of logical qubits mapped to a single physical qubit .
also , we require _ dilation _ , which is the maximum number of stretched edges ( through other qubits ) , to be exactly 1
. however , all of the existing research on embedding problems for parallel processors @xcite , at least one of the conditions is violated ( namely either load @xmath75 or dilation @xmath75 ) . in this paper , our focus is on parameter setting such that the reduction is correct , and not on the minor - embedding algorithm and/or the related ( minor - universal ) hardware graph @xmath1 design .
these problems will be addressed in a subsequent paper .
let the qubo problem , specified by @xmath51 and @xmath0 , be given as in eq . , and the corresponding @xmath39 in eq
.. suppose @xmath2 is a minor - embedding of @xmath0 , and let @xmath76 be the embedded energy function associated with @xmath2 : @xmath77 where @xmath78 .
our goal is to find out the requirement for new parameters , @xmath79s and @xmath80s , such that one can solve the original ising problem on @xmath0 by solving the embedded ising problem on its embedding @xmath2 , or equivalently , such that there is an one - one correspondence between the minimum of @xmath39 ( and thus the maximum of @xmath51 ) and the minimum of @xmath76 .
as we mentioned in the introduction , the idea that we can solve the original ising problem ( i.e. finding the groundstate of the ising hamiltonian ) on the input graph @xmath0 by solving the new ising problem on the embedded graph @xmath2 in @xmath1 is that one can use ferromagnetic couplers to connect the physical qubits in each @xmath66 of @xmath2 such that the subtree @xmath66 will act as one logical qubit @xmath18 of @xmath0 .
[ [ notation . ] ] notation .
+ + + + + + + + + first , we recall that @xmath81 and @xmath82 we distinguish the edges within @xmath66s from the edges corresponding to the edges in the original graph . denote the latter by @xmath83 ,
that is , @xmath84 . ( the black edges in @xmath2 of figure [ fig : example - embedding ] correspond to @xmath83 . ) for convenience , @xmath85 and @xmath86 , @xmath87 .
note @xmath88 .
the above intuition suggests that we use the same coupler strength for each original edge , i.e. , @xmath89 for @xmath90 , and use ferromagnetic coupler strength , @xmath91 , for each edge @xmath92 , and redistribute the bias @xmath20 of a logical qubit @xmath18 to its physical qubits in @xmath66 .
that is , we choose @xmath93 for physical qubit @xmath87 such that @xmath94 .
therefore , we have @xmath95 in this section , we derive an easy upper bound for the ferromagnetic coupler strengths .
the derivation is based on the _ penalty or multiplier _ method , in which a constrained optimization is reduced to an unconstrained one by replacing the corresponding expression into the objective function as a penalty term with a large multiplier .
for example , this was used in the reduction from general ( higher - order ) unconstrained binary optimization to the quadratic one ( see @xcite ) .
notice that by construction ( i.e. , @xmath0 is a minor of @xmath2 and @xmath93s are chosen such that @xmath96 ) , minimization of @xmath39 in is equivalent to @xmath97 where the condition @xmath98 for all @xmath99 is equivalent to requiring the spins of physical qubits that correspond to the same logical qubit @xmath18 to be of the same sign .
the corresponding unconstrained minimization is thus @xmath100 which is equivalent to solving eq .
as @xmath101 is a constant .
we are interested in how large ( in terms of magnitude ) @xmath102s are sufficient to guarantee that the solution of eq .
gives the solution to eq .
, and consequently to eq .. the result is stated in the following theorem .
let @xmath103 and @xmath104 given as in eq .
and eq .. suppose that for @xmath59 , @xmath105 then we have @xmath106 , @xmath107 , where @xmath108 .
consequently , @xmath109 [ thm : easy ] let @xmath110 suppose on the contrary .
that is , there exists @xmath111 , for some @xmath112 such that @xmath113 .
then we have @xmath114 where @xmath115 thus , if @xmath116s satisfy eq . , we have @xmath117 which is at least @xmath118 , a contradiction .
consequently , we can conclude that @xmath119 , @xmath107 , where @xmath108 , implying the claimed equality of the minima .
as we mentioned section [ sec : introduction ] , it is desired to obtain tighter bounds for the ferromagnetic coupler strengths . in this section
, we show that with a more careful analysis , we can reduce the bound by setting the qubit bias ( @xmath79 ) appropriately . for @xmath59 ,
let @xmath120 for @xmath121 , where @xmath122 .
therefore , wlog , for the rest of the paper , we will assume that @xmath123 .
let @xmath39 ( @xmath51 resp . ) and @xmath0 be given as in eq .
( eq . resp . ) .
suppose @xmath2 is a minor - embedding of @xmath0 , and let @xmath76 be the energy of the embedded ising hamiltonian given in eq .. then for all @xmath59 , if @xmath124 and @xmath125 where @xmath123 ( defined above ) and @xmath126 resp . ) if @xmath127 ( @xmath121 resp . )
, we have @xmath106 , @xmath107 , where @xmath128 .
consequently , there is one - one correspondence between @xmath129 and @xmath40 ( and thus @xmath130 ) .
furthermore , if we set @xmath131 , for some @xmath132 , then the spectral gap ( which is the difference between the two lowest energy levels ) of the embedded ising hamiltonian will be the minimum of @xmath133 and the spectral gap of the original ising hamiltonian .
[ thm : main ] see figure [ fig : new - parameters ] for an example of the corresponding parameters in the embedded ising hamiltonian ( for the topological - minor - embedding case ) . in the following ,
we first explain the main idea behind , followed by the formal proof . for the illustration purpose , suppose @xmath134 , consider the simplest case in which all but one leaf is @xmath135 .
now , consider the energy change if the leaf is flipping from @xmath136 to @xmath135 . our goal is to set @xmath116s as small ( in terms of the magnitude ) as possible , such that the energy change is at least greater than zero .
note that the energy change @xmath137 @xmath138 where @xmath139 depends on the sign of spin @xmath140 .
we would like to bound @xmath141 without knowing the signs of @xmath140s .
observe that if @xmath142s are all positive , then the worst case we have @xmath143 . and in general @xmath144 , where @xmath145 and @xmath146 .
consequently , setting @xmath147 would imply @xmath148 .
one can then extend this argument to a segment that needs to be flipped , as illustrated in figure [ fig : sketch - proof ] .
@xmath149{proof - sketch1.eps } & \includegraphics[width=3in]{proof - sketch.eps } \\ ( a ) h>0 : \delta{{\mathcal{e}}}\ge 2 ( h - j^+ - j^+ -f ) & ( b ) h>0 : \delta{{\mathcal{e}}}\ge 2 ( h - j^+ + j^- -f ) \\ \\ \includegraphics[width=3in]{proof - sketch2.eps } & \includegraphics[width=3in]{proof - sketch3.eps } \\ ( c ) h>0 & ( d ) h<0 \end{array}\ ] ] it is easy to check that @xmath150
. we will first prove the theorem when @xmath2 is a topological - minor - embedding of @xmath0 .
that is , each @xmath66 is a chain @xmath151 connected by consecutive vertices .
thus , the corresponding embedded energy function is given by @xmath152 in this case , we have @xmath153 .
we prove by contradiction .
suppose not .
that is , there exists @xmath59 , and @xmath154 such that @xmath155 .
we distinguish two cases based on the value of @xmath20 . for @xmath134 ,
let @xmath156 be the smallest index such that @xmath157 and @xmath158 , and let @xmath159 be the smallest index such that @xmath160 and @xmath161 .
that is , we have @xmath162 , @xmath163 .
note we have at least either @xmath164 or @xmath165 .
( otherwise they are all @xmath136 ) .
then we claim that by flipping the segment @xmath166 from @xmath136s to @xmath135s , the energy decreases , contradicting to the optimality .
( notice that the @xmath116s within the segment do not change . )
the energy change equals to @xmath167 the argument for @xmath121 is similar except that we will flip from @xmath135 to @xmath136 instead , and @xmath168 therefore , in both cases , if @xmath169 , for all @xmath170 , we have @xmath171 , contradicting to the optimality .
hence all @xmath172 must be of the same sign . for the general minor - embedding when @xmath66s are trees
, we can similarly argue that for @xmath134 , if all but one leaf is positive ( @xmath136 ) , then flipping them ( from @xmath136s to @xmath135s ) will decease the energy provided that @xmath173 .
similarly , one can argue for the case when @xmath121 .
notice that if we set @xmath174 , then the energy change @xmath175 . in this case
, the spectral gap of the embedded ising hamiltonian will be the minimum of @xmath133 and the spectral gap of the original ising hamiltonian .
[ [ remark . ] ] remark .
+ + + + + + + one can generally set @xmath176 where @xmath177s are chosen such that @xmath178 , and set the @xmath116s accordingly .
this flexibility is useful when the precision for parameters is limited .
( in the above theorem , we give an upper bound for the ferromagnetic coupler strengths .
a natural question is how tight our bound is .
can one get an better bound without solving the original problem ? )
first , we formulate wmis problem as a special case of qubo in section [ sec : mis - special - qubo ] . then
, we apply theorem [ thm : main ] to set parameters for the embedded ising hamiltonian for the corresponding mis problem in section [ sec : embed - mis ] . [ [ weighted - mis - wmis . ] ] weighted mis ( wmis ) . + + + + + + + + + + + + + + + + + + + + given an undirected vertex - weighted graph @xmath0 .
let @xmath179 , let @xmath180 be the weight of vertex @xmath18 .
wmis seeks to find a @xmath181 such that @xmath182 is independent and the total weight of @xmath182 ( @xmath183 ) is maximized . if @xmath184 for all @xmath68 , then the maximum value of @xmath185 is the total weighted of the wmis . in particular if @xmath186 for all @xmath68 , then @xmath187 , where @xmath188 . [ thm : mis ] let @xmath189 .
denote @xmath190 .
we ll prove that if @xmath186 for all @xmath68 , then @xmath191 is an independent set .
suppose on the contrary , that is , there exists an edge @xmath192 in the subgraph induced by @xmath191 .
wlog , assume that @xmath193 .
consider removing @xmath194 from @xmath195 .
let @xmath196 .
the weight change equals to @xmath197 , contradicting to the optimality of @xmath195 .
the above theorem is a generalization of the known fact for unweighted case of mis ( see @xcite and references therein ) . for the unweighted case of mis , @xmath198 for all @xmath59 .
thus , it is sufficient to choose @xmath199 for all @xmath200 for some @xmath201 .
accordingly , the corresponding energy function of the ising model for mis is @xmath202 .
[ [ remark.-1 ] ] remark .
+ + + + + + + if we choose @xmath142 to be exactly @xmath49 instead ( as in @xcite ) , then we can only guarantee the size of the maximum independent set , but the returned set is not necessarily independent .
for instance , when @xmath203 , any ( adjacent ) two vertices also has the minimum energy of @xmath204 .
note that we thus can conclude that the ising problem is np - hard because wmis is np - hard .
indeed , barahona @xcite showed the np - hardness of a special ising problem through the reduction of mis problem on cubic graph ( which remains np - hard ) . notice that for wmis on a planar graph , there is a ptas algorithm @xcite .
recently , for the ising problem on a planar graph , bansal et al .
@xcite applied the same technique to obtain a ptas algorithm in @xmath205 where @xmath206 is the approximation ratio .
as another side note , conversely ( to the fact that wmis is a special case of qubo ) , boros et al ( see e.g. @xcite ) has shown that a problem of qubo on @xmath0 can also be converted to a wmis but on a different graph .
let @xmath0 be the graph of mis problem and @xmath2 be the topological - minor - embedding of @xmath0 in @xmath1 .
for the unweighted mis , we have @xmath207 .
therefore , according to theorem [ thm : main ] , it suffices to set @xmath208 .
that is , the embedded ising hamiltonian for solving unweighted mis is given as in eq .
, with @xmath209 where @xmath210 , and @xmath211 , @xmath212 . in particular , for degree-3 hardware graph @xmath1 , by setting @xmath213 , and @xmath214 , for some @xmath201 , there are only 6 different parameter values , namely , @xmath215 in eq . .
in this paper , we introduce minor - embedding in aqc . in particular , we show that the np - hard qubo problem can be solved using an adiabatic quantum computer that implements ising spin-1/2 hamiltonians , through minor - embedding reduction .
there are two components to this reduction : embedding and parameter setting . given a minor - embedding ,
we show how to derive the values for the corresponding parameters , in particular , a good upper bound ( in terms of magnitude ) for the ferromagnetic coupler strengths , of the embedded ising hamiltonian such that there is one - one correspondence between the groundstate of the original ising hamiltonian and the one of the embedded ising hamiltonian .
there are many algorithmic problems related to minor - embedding in aqc that remain to be addressed .
in particular , the problems relate to the efficiency of the reduction .
these problems in turn relate to the running time or complexity of quantum adiabatic algorithms . recall that according to the adiabatic theorem , the running time of an adiabatic algorithm depends on the minimum spectral gap of the system hamiltonian , which however might be as hard as solving the original problem . despite several serious investigations ,
the power of aqc remains an open question @xcite .
how does the embedding reduction effect the time complexity of an adiabatic algorithm ? in order to address this question , one will also need to specify the initial hamiltonian . in @xcite
, we show that for some special cases , how the embeddings , parameters , and initial hamiltonians can effect the minimum spectral gaps .
the effect of the embedding and its consequential initial hamiltonian on the complexity of adiabatic algorithms remains to be investigated . in the following we state several main problems that need to investigate in order to facilitate the design of adiabatic algorithms and aqc architectures .
partial results to these problems will appear in our subsequent papers .
[ [ p1.-measurement - for - the - minor - embedding . ] ] p1 .
measurement for the minor - embedding .
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + define a measure for the minor - embedding such that a good minor - embedding corresponds to a reduced problem that admits an efficient adiabatic algorithm .
[ [ p2.-embedding - dependent - initial - hamiltonian . ] ] p2 .
embedding - dependent initial hamiltonian .
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + design an embedding - dependent initial hamiltonian for a given minor - embedding such that the adiabatic algorithm for the reduced problem is at least as efficient as the adiabatic algorithm ( with the best possible initial hamiltonian ) for the original problem .
[ [ p3.-hardware - graph - design . ] ] p3 . hardware graph design .
+ + + + + + + + + + + + + + + + + + + + + + + + + + given a family @xmath216 of graphs ( which consists of classically hard instances ) , the problem is to design a hardware graph ( called a _
@xmath216-minor - universal _ graph ) which is as small as possible ( in terms of total number of vertices and edges ) such that * all known physical constraints are satisfied ; * all graphs in @xmath216 are embeddable ; * a good embedding of each graph in @xmath216 can be efficiently computed .
i would like to thank my incredible colleagues : mohammad amin , andrew berkley , richard harris , mark johnson , jan johannson , andy wan , colin truncik , paul bunyk , felix maibaum , fabian chudak , bill macready , and geordie rose .
thanks also go to david kirkpatrick for the discussion , advice and encouragement .
i would also like to thank bill kaminsky for his detailed comments .
10 d. aharonov , w. van dam , j. kempe , z. landau , s. lloyd , and o. regev .
adiabatic quantum computation is equaivalent to standard quantum computation . , pages 4251 , 2004 .
m. h. s. amin and v. choi .
work in progress .
m. h. s. amin , p. j. love , and c. j. s. truncik .
thermally assisted adiabatic quantum computation .
, 100:060503 , 2008 .
m. h. s. amin , c. j. s. truncik , and d. v. averin .
the role of single qubit decoherence time in adiabatic quantum computation .
. b. s. baker .
approximation algorithms for np - complete problems on planar graphs .
, 41(1):153180 , 1994 .
n. bansal , s. bravyi , and b. m. terhal . a classical approximation scheme for the ground - state energy of ising spin hamiltonians on planar graphs .
f. barahona . on the computational complexity of ising spin glass models . ,
pages 15 : 32413253 , 1982 .
e. boros and p. hammer .
pseudo - boolean optimization . , ( 123):155225 , 2002 .
e. boros , p. l. hammer , and g. tavares .
preprocessing of quadratic unconstrained binary optimization . ,
s. bravyi , d. p. divincenzo , d. loss , and b. m. terhal .
simulation of many - body hamiltonians using perturbation theory with bounded - strength interactions . .
a. childs , e. farhi , and j. preskill .
robustness of adiabatic quantum computation .
, 65 , 2002 .
r. diestel . .
springer - verlag , heidelberg , 2005 .
l. f. thomson .
e. farhi , j. goldstone , s. gutmann , j. lapan , a. lundgren , and d. preda .
a quantum adiabatic evolution algorithm applied to random instances of an np - complete problem .
, 292(5516):472476 , 2001 .
e. farhi , j. goldstone , s. gutmann , and m. sipser .
quantum computation by adiabatic evolution . , 2000 .
l. m. ioannou and m. mosca
. limitations of some simple adiabatic quantum algorithms . , 2007 .
w. m. kaminsky and s. lloyd .
scalable architecture for adiabatic quantum computing of np - hard problems . in a. j. leggett , b. ruggiero , and p. silvestrini , editors , _ quantum computing and quantum bits in mesoscopic systems _ , 2004 .
w. m. kaminsky , s. lloyd , and t. p. orlando .
scalable superconducting architecture for adiabatic quantum computation . ,
j. kempe , a. kitaev , and o. regev .
the complexity of the local hamiltonian problem .
, 35:1070 , 2006 . j. m. kleinberg and r. rubinfeld .
short paths in expander graphs . in _ ieee symposium on foundations of computer science _ , pages 8695 , 1996 .
r. oliveira and b. m. terhal . the complexity of quantum spin systems on a two - dimensional square lattice . .
b. w. reichardt .
the quantum adiabatic optimization algorithm and local minima . in _
stoc 04 : proceedings of the thirty - sixth annual acm symposium on theory of computing _ , pages 502510 , new york , ny , usa , 2004 .
n. robertson and p. d. seymour .
graph minors .
xiii : the disjoint paths problem .
, 63(1):65110 , 1995 .
w. van dam , m. mosca , and u. vazirani .
how powerful is adiabatic quantum computation ? , pages 279287 , 2001
. w. van dam and u. vazirani .
limits on quantum adiabatic optimization . , 2001 .
* appendix *
in this section , we show that the energy function of the ising hamiltonian in eq . is given in eq . .
that is , replace each @xmath217 by the variable @xmath37 .
the equivalence follows from two basics : ( 1 ) eigenvalues and eigenvectors of @xmath27 : @xmath218 , @xmath219 , where @xmath220^{\dagger}$ ] , @xmath221^{\dagger}$ ] .
( 2 ) tensor product property : @xmath222 .
more precisely , for @xmath223 , @xmath224 , @xmath225 . | we show that the np - hard quadratic unconstrained binary optimization ( qubo ) problem on a graph @xmath0 can be solved using an adiabatic quantum computer that implements an ising spin-1/2 hamiltonian , by reduction through _ minor - embedding _ of @xmath0 in the quantum hardware graph @xmath1 .
there are two components to this reduction : _ embedding _ and _ parameter setting_. the embedding problem is to find a minor - embedding @xmath2 of a graph @xmath0 in @xmath1 , which is a subgraph of @xmath1 such that @xmath0 can be obtained from @xmath2 by contracting edges .
the parameter setting problem is to determine the corresponding parameters , qubit biases and coupler strengths , of the embedded ising hamiltonian . in this paper
, we focus on the parameter setting problem . as an example , we demonstrate the embedded ising hamiltonian for solving the maximum independent set ( mis ) problem via adiabatic quantum computation ( aqc ) using an ising spin-1/2 system .
we close by discussing several related algorithmic problems that need to be investigated in order to facilitate the design of adiabatic algorithms and aqc architectures . |
in this paper we present the results of our follow - up work on the 9th cambridge survey ( 9c hereafter ) .
the 9c survey was carried out at 15ghz with the ryle telescope ( rt see * ? ? ?
* ) , primarily motivated by the need to identify foreground sources in the fields surveyed by the cambridge / jodrell / tenerife cosmic microwave background experiment , the very small array ( vsa , e.g. * ? ? ?
a rastering technique was used to scan the fields with each possible detection being followed up with pointed observations to confirm its existence and measure its flux density or to rule it out : see @xcite for a full description of 9c .
the survey fields were chosen to contain few very bright radio sources , but otherwise should be representative of the radio sky in general .
15ghz is the highest radio frequency at which an extensive survey has been carried out , so 9c provides new insights into the properties of the radio source population .
current models of radio source growth that consider the effects of self absorption on the synchrotron emission from young sources indicate that very young radio sources ( tens to hundreds of years old ) should have radio spectra which peak between about one gigahertz and a few tens of gigahertz @xcite , with younger sources having spectra peaking at higher frequencies than older sources .
any radio survey is biased toward the selection of sources with spectra peaking close to the selection frequency , hence 9c should provide a means of generating samples rich in sources peaking at close to 15ghz and thereby testing the models of source growth in very young sources .
we have selected 176 sources from the 9c survey ( 155 of which are from complete flux - limited samples ) and carried out multi - frequency simultaneous radio observations to obtain the continuum radio spectra and maps . _
r - band optical imaging was performed in order to make optical identifications ( ids ) .
_ the layout of this paper is as follows . in [ sec : data ] we discuss sample selection , data acquisition and data reduction . in [ sec : results ] we present the results the radio flux densities , radio maps and optical counterpart data . in [ sec : stats ] we discuss the sample statistics with regard to the radio spectra , radio morphology , optical colour and optical morphology . in [ sec : compare ] we compare these results with previous work and in [ gps ] we consider a means of increasing the efficiency of selecting rising spectrum sources .
we summarise our findings in [ sec : discussion ] .
two complete samples of sources were selected from the first three regions of the 9c survey , coincident with the vsa fields at @xmath0 , @xmath1 and @xmath2(j2000.0 ) .
sample a is complete to 25 mjy and contains 124 sources selected from regions in all three fields , a total area of 176 square degrees .
sample b is complete to 60 mjy , with 70 sources in a total area of 246 square degrees ; it consists of all sources in sample a above 60 mjy ( 39 sources ) , plus 31 additional sources from a further region of the 9h field .
additionally 21 9c sources ( sample c ) were observed which were either outside the sample areas or had flux densities lower than the selection limit
these do not form a complete sample .
simultaneous continuum snapshot observations were made for each source at frequencies of 1.4 , 4.8 , 22 and 43ghz with the vla of the national radio astronomy observatory ( table [ radio_obs ] ) and at 15ghz with the rt .
in addition , 51 sources from the @xmath3 field were observed within a few months at 31ghz with the owens valley radio observatory ( ovro ) 40 m telescope .
the _ uv - plane coverage of the vla differs significantly for the different sets of observations .
the data from jan 2002 were taken in the @xmath4 move configuration and only a few antennas had been moved into their a - array positions ; although at 4.8ghz , for example , there are baselines out to 500kilo-@xmath5 , the majority are less than 15kilo-@xmath5 and the resulting beam is messy . in order to obtain good flux density estimates and a smooth beam , the central portion of the _ uv - plane ( corresponding to the d - configuration baselines ) was used ; after this , the full _ uv - coverage was used to look for structure on smaller angular scales _ _ _ the vla data were reduced using the nrao ` aips ` package . for each dataset maps
were made and cleaned with the ` aips ` task ` imagr ` .
self - calibration was applied to those maps with sufficiently high signal to noise ratio typically sources with point - like components having flux densities of around 40mjy or greater . in each case
one or more rounds of self - calibration , in phase and in amplitude and phase , were carried out to maximise the signal to noise ratio of the final map .
time spent on source was typically about 40 s at 1.4ghz , 60 s at 4.8ghz , 100 s at 22ghz and 120 s at 43ghz , giving typical rms noise values of 0.5 , 0.4 , 0.8 and 2mjy respectively .
the vla flux calibration is assumed to be good to about 1percent , 2percent , 5percent and 10percent at 1.4 , 4.8 , 22 and 43ghz respectively .
each source was observed for about 20 minutes with the rt ; the rms noise is about 0.9mjy and the calibration uncertainties are approximately 3percent .
the ovro 40-m observations were carried out between 2002 jan and 2002 july .
the rms noise on the flux density measurements is typically 1 mjy , but is often higher for the brighter sources .
flux calibration uncertainties are about 5percent . [
cols="<,<,<,<,^,^,^,^,^,^,^ " , ]
we have defined complete , flux - limited samples from a survey at 15ghz that contain higher fractions of gps sources than samples selected at lower radio frequency . future work on gps sources will benefit if complete samples of such objects can be selected without the need for time - consuming multi - frequency measurements of all objects in the flux - limited samples .
the efficiency of selecting gps sources can be increased by removing sources with steeply falling spectra as indicated by existing survey measurements : in particular comparison of the old , non - simultaneous nrao vla sky survey ( nvss , see * ? ? ?
* ) flux density measurements at 1.4ghz and the 9c catalogue flux densities for each object gives an indication of the spectral type .
figure [ c ] shows histograms of the ratio between the simultaneous flux density measurements to those from surveys the top figure shows this histogram for the 1.4-ghz data ( follow - up vla : nvss ) and the lower is for the 15-ghz data ( follow - up rt : 9c ) .
the 15ghz histogram is wider , indicating , as expected , that source variability is more prevalent at 15ghz than at lower radio frequency .
nvss was carried out between 1993 and 1996 and the 9c data points were taken between 1999 november and 2001 june , so the time offset between the survey data points and our simultaneous data points is around 6 years for the 1.4ghz data and between about 1 and 2 years at 15ghz . to quantify the variability of the 9c samples on timescales from a few days to a year or
so we are carrying out systematic rt observations at 15ghz of one third of the sources from the flux - limited samples ( chosen randomly ) , to be presented elsewhere .
figure [ oldlc ] shows the correlation between spectral index calculated between 1.4 and 15ghz from nvss and 9c and the spectral index calculated from the simultaneous measurements at 1.4 and 4.8ghz ( i.e. the spectral index we use to classify the radio sources ) .
the graph shows that the two measurements are similar , but there is significant scatter .
the gps sources with @xmath6 have a tendency for their nvss/9c spectral indices to be higher than their simultaneous 1.4 to 4.8ghz spectral indices .
this is due to the turnover or flattening of the spectra of these sources between 4.8 and 15ghz .
note however that , as suggested in section [ sec : radio_stats ] , the sources have not turned over sufficiently to have steeply falling spectra when averaged between 1.4 and 15ghz . selecting all sources that have an nvss/9c spectral index less steep than @xmath7 will pick up all of the rising spectrum sources , as well as a few flat spectrum sources , but very few steep spectrum objects ( since the spectral index of a convex - spectrum source increases with frequency ) . using the spectral
index cut - off of @xmath8 , and a 9c flux limit of 25mjy creates a sample in which 49percent of the sources would have a simultaneous spectral index of @xmath6 .
our follow - up of 176 radio sources taken from the 9c catalogue has shed light on the properties of the high frequency radio population .
multi - frequency simultaneous measurements have provided maps and radio spectra .
optical imaging at the palomar 60-inch telescope in combination with the dss2 digitised form poss - ii sky survey plates has led to the identification of counterparts for a total of 155 of the 176 sources studied .
( ) at a flux limit of 25mjy , 18percent of sources display spectra that peak at 5ghz or above .
27percent of the 60mjy sample peak at 5ghz or above .
these rising spectrum sources represent a much larger fraction of the radio source population at 15ghz than at lower radio frequencies .
( ) increasing the flux limit raises the fraction of rising spectrum sources : in a sample of sources selected with a 9c flux density at 15ghz of more than 150mjy , 39percent of the sources have rising spectra between 1.4 and 5ghz .
( ) at higher frequencies the median spectral index increases , due to turn - over .
however , the median value remains constant up to 15ghz .
( ) the optical counterparts of the steep spectrum sources are less often blue in colour than those of the flat and rising spectrum sources .
the optical colours and sizes suggest that between one third and one half of the peaked spectrum and flat spectrum sources may be quasars .
( ) we have shown that a sample defined as a flux - limited complete sample from 9c and then refined to contain only those sources having a spectral index taken from nvss at 1.4ghz and 9c at 15ghz of @xmath8 will still contain all of the peaked - spectrum sources from the complete sample .
the efficiency of finding peaked - spectrum objects can thus be raised to 49percent allowing larger samples of gps sources to be studied with the same amount of dedicated telescope time , without compromising on completeness .
rcb is supported by a pparc studentship .
gc acknowledges support from pparc observational rolling grants ppa / g / o/2003/00123 and ppa / g / o/2001/00575 .
the second palomar observatory sky survey ( poss - ii ) was made by the california institute of technology with funds from the national science foundation , the national aeronautics and space administration , the national geographic society , the sloan foundation , the samuel oschin foundation , and the eastman kodak corporation .
the digitized sky survey was produced at the space telescope science institute under u.s .
government grant nag w-2166 .
the images of these surveys are based on photographic data obtained using the oschin schmidt telescope on palomar mountain and the uk schmidt telescope .
the plates were processed into the present compressed digital form with the permission of these institutions .
99 becker r.h . ,
white r.l . ,
helfand d.j . , 1995 , aj , 450 , 559 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 jones m.e . , 1991 , in cornwell t.j .
, perley r. , eds , asp conf .
19 , radio interferometry - theory , techniques and applications .
, san fransisco , p. 395
mclean b.j . ,
greene g.r . ,
lattanzi m.g .
, pirenne b. , 2000 , in manset n. , veillet c. , crabtree d. , eds , asp conf .
216 , astronomical data analysis software and systems ix , p. 145
mcmahon r. , white r. , helfand d. , becker r. , 2002 , apjs , 143 , 1 odea c.p .
, 1998 , pasp , 110 , 493 peacock j.a . , wall j.v . , 1981 , mnras , 194 , 331 reid i.n .
et al . , 1991 , pasp , 103 , 661 riley j.m . , rawlings s. , mcmahon r.g . , blundell k.m . ,
miller p. , lacy m. , waldram e.m .
, 1999 , mnras , 307 , 293 snellen i.a.g .
, schilizzi r.t . ,
miley g.k .
, de bruyn a.g . ,
bremer m.n . ,
rttgering h.j.a . , 2000 , mnras , 319 , 445 waldram e.m . , pooley g.g . ,
grainge k.j.b .
, jones m.e . , saunders r.d.e . ,
scott p.f . ,
taylor a.c . , 2003 , mnras , 342 , 915 watson r.a .
. , 2003 , mnras , 341 , 1057 | we have carried out extensive radio and optical follow - up of 176 sources from the 15ghz 9th cambridge survey .
optical identifications have been found for 155 of the radio sources ; optical images are given with radio maps overlaid .
the continuum radio spectrum of each source spanning the frequency range 1.4 - 43ghz is also given .
two flux - limited samples are defined , one containing 124 sources complete to 25mjy and one of 70 sources complete to 60mjy .
between one fifth and one quarter of sources from these flux - limited samples display convex radio spectra , rising between 1.4 and 4.8ghz .
these rising - spectrum sources make up a much larger fraction of the radio source population at this high selection frequency than in lower frequency surveys .
we find that by using non - simultaneous survey flux density measurements at 1.4 and 15ghz to remove steep spectrum objects , the efficiency of selecting objects with spectra rising between 1.4 and 4.8ghz ( as seen in simultaneous measurements ) can be raised to 49percent without compromising the completeness of the rising spectrum sample .
[ firstpage ] galaxies : active radio continuum : general surveys |
supermassive black holes with masses @xmath8 are ubiquitous in the nuclei of local galaxies of moderate to high mass ( e.g. , * ? ? ?
it is now well established that most of the total mass in black holes in the nearby universe was accreted in luminous episodes with high eddington rates ( e.g. , * ? ? ?
* ; * ? ? ?
* ) , with the growth for massive ( @xmath9 ) black holes occurring predominantly at @xmath10 ( e.g. , * ? ? ?
* ; * ? ? ?
these rapidly accreting black holes are most readily identified as bright optical quasars with characteristic broad ( @xmath11 km s@xmath12 ) emission lines , and luminous continuum emission that can dominate the light from the host galaxy , particularly at ultraviolet and optical wavelengths ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
optical quasars thus provide powerful tools for tracing the rapid growth of black holes over cosmic time ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) . however , it is increasingly clear that a significant fraction of the quasar population does not show characteristic blue continua or broad lines because their nuclear emission regions are obscured .
key evidence for the existence of obscured ( type 2 ) quasars comes from synthesis models of the cosmic x - ray background ( e.g. , * ? ? ?
* ; * ? ? ?
* ) , as well as direct identification of these objects through various observational techniques .
these include selection of luminous quasars with only narrow optical lines @xcite or relatively weak x - ray emission @xcite , detection of powerful radio galaxies lacking strong nuclear optical continua or broad lines ( e.g. , * ? ? ?
* ; * ? ? ?
* ) , and detection of x - ray sources that are optically faint ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , have hard x - ray spectra ( e.g. , * ? ? ?
* ) , or have radio bright , optically weak counterparts ( e.g. , * ? ? ?
* ) . with the launch of the _ spitzer space telescope _
, large numbers of obscured quasars can now be efficiently identified based on their characteristic ( roughly power - law ) spectral energy distributions ( seds ) at mid - infrared ( mid - ir ) wavelengths ( @xmath2324 ) . because mid - ir emission is less strongly affected by dust extinction than optical and ultraviolet light
, obscured quasars can appear similar to their unobscured counterparts in the mid - ir , but have optical emission characteristic of their host galaxies .
a number of studies using mid - ir colors ( * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* hereafter h07 ) , sed fitting @xcite , or selecting objects based on similarities to mid - ir quasar templates ( e.g. , * ? ?
* ) have been successful in identifying large numbers of dust - obscured quasars , indicating that a large fraction , and possibly a majority of rapid black hole growth is obscured by dust .
these large new samples enable detailed statistical studies that can explore the role of obscured quasars in galaxy and black hole evolution . at present
there are a number of possible physical scenarios for obscured quasars ; in the simplest `` unified models '' , obscuration is attributed to a broadly axisymmetric `` torus '' of dust that is part of the central engine , so obscuration is entirely an orientation effect ( e.g. , * ? ? ?
* ; * ? ? ?
alternatively , obscuration may not be due to a central `` torus '' but to larger dust structures such as those predicted during major mergers of galaxies ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , and obscured quasars may represent an early evolutionary phase when the growing black hole can not produce a high enough accretion luminosity to expel the surrounding material ( e.g. , * ? ? ?
* ; * ? ? ?
observations have revealed evidence for obscuration by a `` torus '' in some cases and by galactic - scale structures in others ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , and while there are examples of obscured quasars that show clear signs of radiative feedback on interstellar gas , it is unclear whether they are driving the galaxy - scale outflows invoked in evolutionary models @xcite .
thus the physical nature of obscured quasars remains poorly understood , and analyses with large samples of mid - ir selected quasars will be essential for a more complete understanding of rapidly growing , obscured black holes .
one particularly powerful observational tool is spatial clustering , which allows us to measure the masses of the dark matter halos in which quasars reside .
clustering studies of unobscured quasars have shown that the masses of quasar host halos are remarkably constant with cosmic time , with @xmath13 @xmath14 @xmath15 over the large redshift range @xmath16 ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
this lack of variation in halo mass implies that the bias factor ( clustering relative to the underlying dark matter ) is an increasing function of redshift , since the dark matter is more weakly clustered earlier in cosmic time .
the characteristic @xmath17 provides a strong constraint on models of quasar fueling by the major mergers of gas - rich galaxies ( e.g. , * ? ? ?
* ; * ? ? ?
* ) , secular instabilities ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) or accretion of recycled cold gas from evolved stars @xcite , and may be related to quasars role in regulating star formation and the emergence of the red galaxy population in halos of roughly similar mass @xmath7@xmath18@xmath19 @xmath14 @xmath15 ( e.g. , @xcite ) . despite the power of clustering measurements in understanding quasar populations ,
little is known about the clustering of obscured quasars .
some measurements of lower - luminosity agns indicate no significant difference between obscured and unobscured sources @xcite .
however , these agns likely have different physical drivers compared to powerful quasars ( e.g. , * ? ? ?
* ) . for obscured quasars at high luminosities ( @xmath20 erg s@xmath12 ) and high redshift ( @xmath21 )
, the clustering has remained largely unexplored . in this paper
we present the first measurement of the clustering of mid - ir selected obscured quasars and make direct comparisons to their unobscured counterparts .
we use a large sample of quasars ( both obscured and unobscured ) in the redshift range @xmath0 selected on the basis of irac colors by , using data from the 9 deg@xmath1 botes multiwavelength survey .
we also employ a sample of @xmath2250,000 galaxies with good estimates of photometric redshift , and measure the two - point cross - correlation between quasars and galaxies .
we utilize a novel method developed by ( * ? ? ?
* hereafter m09 ) to derive the projected real - space projected cross - correlation function , making use of the full probability distributions for the photometric redshifts . throughout this paper
we assume a cosmology with @xmath22 and @xmath23 . for direct comparison with other works ,
we assume @xmath24 km s@xmath12 mpc@xmath12 ( except for comoving distances and dark matter halo masses , which are explicitly given in terms of @xmath25 km s@xmath12 mpc@xmath26 ) . in order to easily compare to
estimated halo masses in other recent works on quasar clustering ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , we assume a normalization for the matter power spectrum of @xmath27 .
photometry is presented in vega magnitudes .
all quoted uncertainties are @xmath28 ( 68% confidence ) .
the 9 deg@xmath29 survey region in botes covered by the noao deep wide - field survey ( ndwfs ; * ? ? ?
* ) is unique among extragalactic multiwavelength surveys in its wide field and uniform coverage using space- and ground - based observatories .
extensive optical spectroscopy makes this field especially well suited for studying the statistical properties of a large number of agns ( c. kochanek et al .
2011 , in preparation ) .
further details of the botes data set have been presented in previous papers ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
redshifts for this study come from the agn and galaxy evolution survey ( ages ; kochanek et al .
2011 , in preparation ) which used the hectospec multifiber spectrograph on the mmt @xcite .
we use ages data release 2 ( dr 2 ) , which includes all the ages spectra taken in 20042006 .
details of the agn redshifts are given in and @xcite .
optical photometry from ndwfs was used for the selection of ages targets and to derive optical colors and fluxes for ages sources .
ndwfs images were obtained with the mosaic-1 camera on the 4-m mayall telescope at kitt peak national observatory , with 50% completeness limits of 26.7 , 25.0 , and 24.9 mag , in the @xmath30 , @xmath31 , and @xmath32 bands , respectively .
photometry is derived using sextractor @xcite .
mid - infrared observations are taken from the _ spitzer _ irac shallow survey ( iss ; * ? ? ?
* ) , and _ spitzer
_ deep wide - field survey ( sdwfs ; * ? ? ? * ) .
iss covers the full ages field in all four irac bands ( 3.6 , 4.5 , 5.8 , and 8 @xmath33 m ) , with @xmath34 flux limits of 6.4 , 8.8 , 51 and 50 @xmath33jy respectively .
the irac photometry for iss is described in detail in @xcite .
the more recent sdwfs exposures extend these limits to 3.5 , 5.3 , 30 , and 30 @xmath33jy , respectively .
as discussed below , the quasar sample was selected using iss data , while the galaxy sample for cross - correlation is selected from the full sdwfs data set . in computing bolometric luminosities for the quasars ,
we also make use of 24 flux measurements available from the multiband imaging photometer for _ spitzer _ ( mips ) gto observations ( irs gto team , j. houck ( pi ) , and m. rieke ) of the botes field . significant fluxes ( @xmath35 ) were obtained for 97% of the quasars in our sample that lie in the region covered by mips .
our primary analysis is the two - point cross - correlation between mid - ir selected quasars and galaxies . in this section
we give details of the the quasar ( both obscured and unobscured ) and galaxy samples .
the quasars are taken from the sample of luminous mid - ir selected agns presented by .
quasars are identified on the basis of their colors in the mid - ir as observed by _
irac , using the color - color criterion of @xcite ( figure [ fstern]a ) , and are selected such that their best estimates of redshift are at @xmath36 . to the relatively shallow flux limits of the irac shallow survey ,
the agn sample is highly complete and suffers little contamination from star - forming galaxies ( as discussed in detail in 7 of ; see also @xcite ) .
we note that while refers to the sample as `` agns '' , their bolometric luminosities are estimated to be in the range @xmath37@xmath38 erg s@xmath12 , corresponding roughly to an x - ray luminosity range @xmath39 erg s@xmath12 @xcite .
such high luminosity agns are typically referred to as `` quasars '' in the literature , so to avoid confusion with studies of lower - luminosity active galaxies , here we refer to our sample as `` quasars ''
. showed that at the iss flux limits , the ir - selected quasars show a bimodal distribution in optical to mid - ir color .
the selection boundary at @xmath40=6.1 $ ] can be interpreted as dividing quasars into unobscured ( optically bright and so `` blue '' in @xmath40 $ ] ) and obscured ( optically faint and so `` red '' in @xmath40 $ ] ) subsets ( figure [ fstern]b ) . for the purposes of this study these objects
will be referred to as `` qso-1s '' and `` obs - qsos '' , respectively ; the reader is reminded that the selection is based not on optical spectroscopy but only on optical to mid - ir color .
this selection yields samples of 839 qso-1s and 640 obs - qsos at @xmath36 .
a detailed study of the optical colors , morphologies , and average x - ray spectra of these objects is given in .
to briefly summarize , found that the qso-1s have blue optical colors , point - like optical morphologies , and soft x - ray spectra characteristic of unobscured quasars , while the obs - qsos had redder optical colors , extended optical morphologies , hard x - ray spectra and high @xmath41 characteristic of obscured quasars .
the sample does not include all obscured quasars , as sources with very large extinction may fall below the ir flux limits of the survey or move out of the @xcite selection region ( as shown in figure 1 of ; see also @xcite ) .
the typical absorbing column for the obs - qso sample is estimated to be @xmath42@xmath43 @xmath44 .
we expect the obs - qsos to suffer little contamination from bright star - forming galaxies
. used an x - ray stacking analysis and constraints from deeper surveys to estimate the possible contamination , and concluded that the contamination is at most @xmath230% , and likely significantly smaller ( @xmath4510% ) . for our spatial correlation analysis , we limit the ir - selected quasar sample to the redshift range @xmath0 , to maximize overlap with the normal galaxies in the field ( [ galaxy ] ) .
we also include only objects in regions of good optical photometry and away from bright stars .
these criteria yield 563 qso-1s and 361 obs - qsos . finally , we restrict the qso-1 sample to those spectroscopically identified as broad - line agns , to ensure that they unambiguously represent a sample of unobscured quasars and to enable clean tests of photo-@xmath6 errors ( see [ qsophotoz ] ) . of the full sample of qso-1s all redshifts , the vast majority ( 80% ) have optical spectra from ages and 96% of these
are classified as broad - line agns at @xmath46 , supporting their selection as unobscured quasars .
we limit the qso-1 sample to the 445 that have accurate optical spectroscopic redshifts in the range @xmath0 and clear broad emission line features .
( in a sense this is conservative ; we verify that including the 20% of objects with only photo-@xmath6s has no significant effect on the clustering results . ) based on these selection criteria , our qso-1 sample is essentially equivalent to other type 1 quasar samples selected purely on optical photometric colors and/or spectroscopy ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , since the vast majority of spectroscopic type 1 quasars show agn - like mid - ir colors @xcite .
the positions on the sky of the final samples of qso-1s and obs - qsos are shown in figure [ fagnsky](a ) , and their distribution in redshift is given in figure [ fz ] .
the obs - qsos are ( by definition ) optically faint , and so few ( only 7% ) are bright enough to obtain good redshifts from mmt optical spectroscopy .
ages targeted objects down to a flux limit of @xmath47 for sources that are optically extended , which is the case for almost all the obs - qsos .
therefore the vast majority of the obs - qso sample has only photometric estimates of redshift , derived using an artificial neural net technique @xcite .
uncertainties on photo-@xmath6s using this technique for optically - bright quasars are typically @xmath48 .
however the errors are more difficult to estimate for optically - faint obs - qsos , for which there are few spectroscopic redshifts for comparison .
photo-@xmath6 uncertainties were discussed at length by , with the conclusion that typical uncertainties are at most @xmath49 and are likely smaller .
figure [ fspz ] shows the photo-@xmath6s and spec-@xmath6s for the handful of obs - qsos with spectroscopic redshifts , as well as those for the qso-1s for comparison .
the impact of photo-@xmath6 errors in the present clustering analysis are addressed in detail in [ qsophotoz ] .
as discussed in
[ qsophotoz ] , random errors in the photo-@xmath6s can only tend to _ decrease _ the observed clustering amplitude , so we expect the present analysis to provide a robust lower limit on the clustering of the obs - qsos . since the primary aim of this analysis is to compare the clustering of quasars with and without obscuration by dust
, it is imperative that the samples are otherwise matched in key properties such as redshift and luminosity .
we show in figure [ fz ] that the redshift distributions of the two samples are similar , and we obtain bolometric luminosities ( @xmath50 ) for the quasars by scaling from the rest - frame 8 luminosity .
we compute the flux at rest - frame 8 by extrapolating between the fluxes at 8 and 24 in the observed frame , and use this flux to obtain the monochromatic luminosity @xmath51 at 8 .
we then multiply by a luminosity - dependent bolometric correction from @xcite , which ranges from factors of @xmath28 to 11 , in order to obtain @xmath50 .
visual inspection of the _ spitzer _ data shows that essentially all of the quasars have broadly power - law seds at these wavelengths , indicating the rest - frame 8 emission is indeed dominated by the agn .
we note that 49 quasars lie outside the region covered the mips 24 observations , while 26 ( @xmath23% ) of those inside the mips area are not detected at 24 . for these 75 objects
, we use the estimates of @xmath50 derived from the rest - frame 2 luminosity as in
4.6 of . in general ,
the @xmath50 derived from the rest - frame 2 luminosity as used in ( which did not make use of the 24 data ) broadly matches that obtain from the extrapolated 8 flux .
however , the median @xmath50 obtained from 2 is smaller for the obs - qsos than for the qso-1s by @xmath20.15 dex , primarily because the obs - qsos have somewhat redder mid - ir seds consistent with the nuclear emission being reddened by dust ( e.g. , * ? ? ?
the distributions in @xmath50 are almost identical for the qso-1 and obs - qso samples , as shown in the top panel of figure [ fstern](b ) .
the median and dispersion in @xmath52 ( erg s@xmath12 ) is ( 45.86 , 0.37 ) and ( 45.83 , 0.39 ) for qso-1s and obs - qsos , respectively , indicating that the two samples are very well matched in bolometric luminosity . for completeness
, we note that if we use the @xmath50 estimates derived from rest - frame 2 in and restrict our analysis to qso-1 and obs - qso samples that are matched in @xmath50 , this has a negligible effect on the clustering results .
the sample of 256,124 galaxies is selected from the deeper sdwfs irac observations , with a flux limit @xmath53 < 18.6 $ ] .
the galaxies are selected to have best estimates of photometric redshift between 0.5 and 2 , with an average photo-@xmath6 of @xmath54 .
the sample includes an optical magnitude cut of @xmath55 to restrict it to optical fluxes for which the photo-@xmath6s are well - calibrated .
to eliminate powerful agns , we have also excluded any object detected in 5 ks _
chandra _
x - ray observations @xcite or with 5@xmath56 sdwfs detections in all four irac bands and colors in the @xcite agn selection region .
the exclusion of agns from the galaxy sample removes only 6,979 objects and has negligible effect on the results .
the distribution on the sky of the 256,124 galaxies are shown in figure [ fagnsky](b ) , and their distribution in photometric redshift is shown in figure [ fz ] .
photometric redshifts are obtained using an updated version of the @xcite algorithm , which is based on template fitting to the optical - ir seds .
the sed fitting produces a redshift probability density function ( pdf ) for each object , where @xmath57 represents the probability that the object lies at redshift @xmath6 .
( note that the neural net used for the quasar photo-@xmath6s does not produce an equivalent estimate of the pdf .
thus for the quasars we use the best value for the redshift , as discussed in [ corranal ] .
) @xmath57 is normalized such that @xmath58 . for most galaxies
the pdf is roughly gaussian in shape , although often with a broader tail toward higher redshift .
the typical redshift uncertainties are @xmath59 , and only a small fraction ( @xmath20.6% ) of galaxies show multiple significant peaks in the pdf at different redshifts .
typical galaxy pdfs are shown in figure [ fpdf ] .
in addition to the observed galaxy catalog , the correlation analysis requires a reference sample of objects with random sky positions , in order to compare the observed quasar - galaxy pair counts with the number expected for an uncorrelated distribution .
we use a catalog of @xmath60 random `` galaxies '' that are assigned to random positions in the regions of good photometry , reflecting the spatial selection function for the sdwfs galaxies .
in this section we outline our methods for measuring the spatial cross - correlation between quasars and galaxies , the autocorrelation of the galaxies , and the absolute bias and characteristic dark matter halo masses . to measure the spatial clustering of quasars , we can in principle derive the autocorrelation of the quasars themselves , or measure their _
cross_-correlation with a sample of other objects ( specifically , normal galaxies ) at the same redshifts .
our quasar sample is too small to obtain sufficiently good measurements of their autocorrelation function .
however , cross - correlation with galaxies ( of which there are @xmath61 times as many objects in the botes data set ) allows far greater statistical power .
further , cross - correlation requires knowledge only of the selection function for the galaxies , which is generally better constrained than that for agns .
cross - correlations of agns with galaxies have proved an effective technique in a number of previous studies ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
? * ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
for the present analysis , the uncertainties in the galaxy photo-@xmath6s restrict our ability to perform a full three - dimensional clustering analysis .
however , making use of the quasar redshifts and the galaxy photo-@xmath6 information , we can derive a projected spatial correlation function ( @xmath3 , with @xmath31 in comoving @xmath14 mpc ) that has both higher signal - to - noise , and a more straightforward physical interpretation than , for example , the purely angular correlation function @xmath62 .
the two - point correlation function @xmath63 is defined as the probability above poisson of finding a galaxy in a volume element @xmath64 at a physical separation @xmath65 from another randomly chosen galaxy , such that @xmath66dv,\ ] ] where @xmath67 is the mean space density of the galaxies in the sample .
the projected correlation function @xmath3 is defined as the integral of @xmath63 along the line of sight , @xmath68 where @xmath31 and @xmath69 are the projected comoving separations between galaxies in the directions perpendicular and parallel , respectively , to the mean line of sight from the observer to the two galaxies . by integrating along the line of sight
, we eliminate redshift - space distortions owing to the peculiar motions of galaxies , which distort the line - of - sight distances measured from redshifts .
@xmath3 has been used to measure correlations in a number of surveys , for example sdss @xcite , 2slaq @xcite , deep2 @xcite , botes @xcite , cosmos @xcite and goods @xcite . in the range of separations @xmath70
@xmath14 mpc , @xmath63 for galaxies and quasars is roughly observed to be a power - law , @xmath71 for sufficiently large @xmath72 such that we average over all line - of - sight peculiar velocities , @xmath3 can be directly related to @xmath63 ( for a power law parameterization ) by @xmath73}{\gamma(\gamma/2)}. \label{eqnwpplaw}\ ] ] we use equation ( [ eqnwpplaw ] ) to obtain power - law parameters for the observed correlation functions , to facilitate straightforward comparisons to other works .
however , we note that a number of recent studies have shown evidence for separate terms in the correlation function owing to pairs of galaxies found within a single dark matter halo ( the `` one - halo '' term ) , and from pairs in which each galaxy is in a different halo ( the two - halo term ; e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
a halo occupation distribution ( hod ) analysis accounting for both the one- and two - halo terms can provide valuable constraints on the distribution of objects within their dark matter halos , however a full hod calculation is beyond the scope of the present analysis . to measure @xmath3 for the quasar - galaxy cross - correlation
, we employ the method developed by .
this technique makes use of the full photo-@xmath6 pdf for every galaxy , to weight quasar - galaxy pairs based on the probability of their being associated in redshift space .
we describe the formalism briefly here , and refer the reader to for further details . for a set of spectroscopic quasars all at the same comoving distance @xmath74 from the observer ,
the angular cross - correlation between the ( spectroscopic ) quasars and ( photometric ) galaxies can be expressed in terms of the physical transverse comoving distance by ( e.g. , * ? ? ?
* ) : @xmath75 where @xmath31 is the projected comoving distance for a given angular separation @xmath76 , such that @xmath77 .
@xmath78 and @xmath79 are the total numbers of photometric galaxies and random galaxies , respectively , and @xmath80 and @xmath81 are the number of quasar - galaxy and quasar - random pairs in each bin of @xmath31 . defining the radial distribution function for the full galaxy sample as @xmath82 , where @xmath83 , and assuming that @xmath82 varies slowly at the redshifts of interest , then the angular correlation function @xmath84 is related to the projected real space correlation function @xmath3 by @xmath85 ( for a derivation see 3.2 of @xcite ) .
as discussed in detail in , we can generalize the analysis such that the contribution to @xmath3 is calculated individually for each quasar - galaxy pair , with @xmath86 defined as the average value of the radial pdf @xmath87 for each photometric object @xmath88 , in a window of size @xmath89 around the comoving distance to each spectroscopic object @xmath90 .
we use @xmath91 @xmath14 mpc to effectively eliminate redshift space distortions , although the results are insensitive to the details of this choice . in this case of weighting by pairs ,
we obtain , as in equation ( 13 ) of : @xmath92 where @xmath93
we refer the reader to 2 of for a detailed derivation and discussion of these equations .
use equation ( [ eqnwp ] ) to compute the cross - correlation between spectroscopic and photometric quasars from sdss in the relatively narrow redshift bin @xmath94 , corresponding to comoving distances @xmath95 @xmath14 mpc ; they obtain a cross - correlation length @xmath96 @xmath14 mpc , assuming @xmath97 .
( note that do not derive @xmath97 , they assume it purely to describe their method .
higher values of gamma are typically obtained in the recent literature , and we obtain @xmath98 in the present work . ) our quasar sample spans a comparatively larger range in redshift ( @xmath99 , corresponding to @xmath100 @xmath14 mpc ) .
we evaluate equation ( [ eqnwp ] ) by calculating the @xmath101 term individually for each quasar .
that is , for each quasar and each bin in separation @xmath31 , we sum the redshift weights @xmath102 for galaxies in the given range of distance from the quasar , and divide by the number of random galaxies in the same distance range ( note that this implies @xmath103 in equation [ eqnwp ] ) .
the advantage of this procedure is that it consists of a simple sum and accounts exactly for the comoving distance to each quasar .
however , the calculation is limited by shot noise on small scales where we have small numbers of quasar - galaxy and quasar - random pairs . to check that this does not significantly affect the results
, we also divide the quasar sample into bins of width @xmath104 ( over which the comoving distance variations are small enough that there is little mixing between bins in @xmath31 ) , and calculate the @xmath105 term for all the @xmath106 quasars in the bin .
we then average the @xmath3 values for the different bins to obtain a mean @xmath3 over the redshift range of interest .
the resulting clustering amplitude differs by @xmath10710% ( and in the majority of cases , @xmath45 a few percent ) compared to evaluating equation ( [ eqnwp ] ) treating each quasar separately .
the choice of method does not affect any of our conclusions , but to account for these differences we conservatively include an additional 10% systematic uncertainty on the measurement of the clustering amplitude .
finally , we emphasize that we are averaging @xmath3 over the whole redshift range of @xmath0 .
the validity of this procedure depends on the fact that the observed @xmath3 varies slowly in the redshift range of interest , which we verify explicitly in [ variationz ] . to estimate dark matter halo masses for the quasars , we calculate the relative bias between quasars and galaxies from which we derive the absolute bias of the quasars relative to dark matter . as discussed below , calculation of absolute bias ( and thus halo mass ) requires a measurement of the autocorrelation function of the sdwfs galaxies .
the large sample size enables us to derive the clustering of the galaxies accurately from the angular autocorrelation function @xmath108 alone .
although we expect the photometric redshifts for the sdwfs galaxies to be well - constrained ( as discussed in [ galaxy ] ) , by using the angular correlation function we minimize any uncertainties relating to individual galaxy photo-@xmath6s for this part of the analysis .
the resulting clustering measured for the galaxies has much smaller uncertainties than that for the quasar - galaxy cross - correlation . to save computation time ,
for the galaxy autocorrelation analysis we use a significantly smaller random catalog with only @xmath109 random `` galaxies '' .
this likely introduces some additional shot noise into the calculation of @xmath108 , however since the resulting uncertainties are still far smaller than those for the quasar - galaxy cross - correlation , they are more than sufficient for the present analysis .
we calculate the angular autocorrelation function @xmath108 using the @xcite estimator : @xmath110 where @xmath111 , @xmath112 , and @xmath113 are the number of data - data , data - random , and random - random galaxy pairs , respectively , at a separation @xmath76 , where each term is scaled according to the total numbers of quasars , galaxies , and randoms .
the galaxy autocorrelation varies with redshift , owing to the evolution of large scale structure , and because the use of a flux - limited sample means we select more luminous galaxies at higher @xmath6 .
this will affect the measurements of relative bias between quasars and galaxies , since the redshift distribution of the quasars peaks at higher @xmath6 than that for the galaxies and so relatively higher-@xmath6 galaxies dominate the cross - correlation signal . to account for this in our measurement of galaxy autocorrelation , we randomly select galaxies based on the overlap of the pdfs with the quasars in comoving distance ( in the formalism of [ crosscorr ] this is @xmath86 for each galaxy , averaged all quasars ) .
we select the galaxies so their distribution in redshift is equivalent to the _ weighted _ distribution for all galaxies ( weighted by @xmath114 ) .
the redshift distribution of this galaxy sample is shown in figure [ fgalselect ] .
we use this smaller galaxy sample to calculate the angular autocorrelation of sdwfs galaxies . in fields of finite size , estimators of the correlation function based on pair counts are subject to the integral constraint , which can be expressed as @xcite @xmath115 where @xmath116 is the angle between the solid angle elements @xmath117 and @xmath118 and the integrals are over the survey area .
if the number density fluctuations in the volume are small , and the angular correlations are smaller than the variance within the volume , then to first order the correlation function is simply biased low by a constant equal to the fractional variance of the number counts .
a straightforward way to remove this bias is to add to the observed @xmath108 the term @xmath119 where @xmath120 is the area of the survey region .
the value of @xmath121 , where @xmath122 is an estimate of the mean number of galaxies per unit area , is the contribution of clustering to the variance of the galaxy number counts @xcite .
evaluating equation ( [ eqnint ] ) for the botes survey area and the typical slope of the @xmath108 for the objects considered here , we obtain @xmath123 .
we estimate @xmath124 by interpolating the observed @xmath108 , then add @xmath125 to @xmath108 before performing model fits . for the projected real - space correlation functions @xmath3 ( which is ultimately derived from individual estimates of @xmath108 , as in equation [ eqnccorr ] )
, we perform an approximate correction for the integral constraint .
we determine the value of @xmath3 at the physical scale ( typically 0.51 @xmath14 mpc ) corresponding to 1 for each quasar , and add the average of these estimates ( multiplied by 0.03 ) to the observed @xmath3 .
these corrections increase the observed clustering amplitude by @xmath210% , but have little effect on our overall conclusions .
ideally , uncertainties in @xmath3 and @xmath108 would be determined by calculating the correlation function for various random realizations of mock ir - selected quasar and galaxy samples , for example by populating dark matter @xmath126-body simulations . in the absence of such mock catalog ,
we instead determine uncertainties in @xmath3 directly from the data through bootstrap resampling . in a standard bootstrap analysis ,
the survey volume is divided into @xmath127 subvolumes , and these subvolumes are drawn randomly ( with replacement ) for inclusion in the calculation of the correlation function .
owing to the relatively small size of the field compared to large surveys such as sdss or 2df , we are only able to divide the field into a small number of subvolumes ( we choose @xmath128 ) .
the width of one subvolume corresponds to @xmath250 @xmath14 mpc at @xmath129 , so that correlations between the subvolumes should be relatively weak .
( we verify explicitly that using a larger @xmath130 has no significant effect on the results . ) for each bootstrap sample draw a total of @xmath131 subvolumes ( with replacement ) , which has been shown to best approximate the intrinsic uncertainties in the clustering amplitude @xcite .
we then re - calculate @xmath3 including only the subvolumes in the bootstrap sample . for the calculations of @xmath3 we use 10,000 bootstrap samples , for which the uncertainties at each scale converge to better than 1% .
( to save computing time , we limit the analysis to 2000 bootstrap samples for the angular correlation analyses , for which the uncertainties converge to within @xmath21.5% . ) this bootstrap technique works well for the galaxy autocorrelation , for which we have a large number of objects and the uncertainty is dominated by the clustering of the sample rather than counting statistics . however , for the quasar - galaxy cross - correlation the bootstrap analysis results in very small errors that are significantly smaller than the observed scatter between points .
this appears to be caused by the fact that , owing to the small quasar samples of only a few hundred objects , the uncertainties are dominated by shot noise that is not fully characterized by randomly selecting entire subvolumes . to account for the shot noise
, we therefore take the sets of @xmath131 bootstrap subvolumes and randomly draw from them ( with replacement ) a sample of objects ( quasars or galaxies ) equal in size to the parent sample ; only pairs including these objects are used in resulting cross - correlation calculation .
this procedure yields a good estimate of the shot noise ( the resulting @xmath132 ) while also accounting for covariance due to the large - scale structure .
when fitting power - law models to the observed correlation functions , we compute parameters by minimizing @xmath133 , taking into account covariance between different bins in @xmath31 . from the bootstrap analysis , we can estimate the covariance matrix @xmath134 by @xmath135 \ , \end{aligned}\ ] ] where @xmath136 and @xmath137 are the projected correlation function derived for the @xmath138-th bootstrap samples , @xmath126 is the total number of bootstrap samples , and @xmath3 is the correlation function for the full sample .
this formalism is equally valid for bins of angular separation @xmath76 in calculations of @xmath108 .
the 1@xmath56 uncertainty in each bin in @xmath31 is the square root of the diagonal component of this matrix ( @xmath139 ) .
taking into account covariance , @xmath133 is defined as @xmath140 where @xmath141 is the inverse of the covariance matrix @xmath134 .
we determine best - fit parameters by minimizing @xmath133 , and derive 1@xmath56 errors in each parameter by the range for which @xmath142 . as a check , we also estimate parameter uncertainties by calculating best - fit parameters for each of the bootstrap samples and calculating the variance between them ; this obtains almost identical estimates of the errors .
further , we note that if we use only the diagonal terms in the covariance matrix in determining @xmath133 , the variation in the best - fit parameters is significantly smaller than the statistical uncertainties , indicating that the precise details of the covariance matrix are relatively unimportant .
we also note that while in principle the sdwfs field is large enough to enable measurements of clustering up to @xmath750 @xmath14 mpc at @xmath143 , we limit the analysis to scales @xmath144 @xmath14 mpc , because of edge effects that skew the correlation function on large scales but have minimal effect on smaller scales .
an investigation of this effect is given in the appendix .
for the projected real - space quasar - galaxy cross - correlation analysis , we fit power - law models of @xmath3 using equation ( [ eqnwpplaw ] ) .
we also fit power laws to the angular correlation functions ( both galaxy autocorrelations and quasar - galaxy cross - correlations ) , using the simple expression @xmath145 for meaningful comparison to other clustering measurements obtained using samples with different distributions in redshift , we wish to convert the observed parameters @xmath146 and @xmath147 to the real - space @xmath148 and @xmath149 as defined in equation ( [ eqnplaw ] ) .
inverting limber s equation , the conversion between these parameters can be computed analytically ( here we follow 4.2 of @xcite ; for the full derivation see @xcite ) : @xmath150\left[\int^{\infty}_0({\rm d}n_2/{\rm d}z){\rm d}z\right ] } r_{0}^\gamma \label{eqndeproj}\end{aligned}\ ] ] where @xmath151\right)/\gamma(0.5\gamma)$ ] , @xmath152 is the gamma function , @xmath153 is the radial comoving distance , @xmath154 are the redshift distributions of the samples ( for an autocorrelation @xmath155 ) , and @xmath156 .
the hubble parameter @xmath157 can be found via @xmath158 . \label{eqnhz}\ ] ] equation ( [ eqndeproj ] ) assumes no evolution with redshift in the clustering of the sample ( equivalent to the implicit assumption made in fitting @xmath3 with equation [ eqnwpplaw ] ) .
for each angular correlation analysis , we derive @xmath146 and @xmath147 from the observed @xmath108 and then obtain the corresponding @xmath149 and @xmath148 from equations ( [ eqndelta ] ) and ( [ eqndeproj ] ) .
the masses of the dark matter halos in which galaxies and quasars reside are reflected in their absolute clustering bias relative to the dark matter distribution . to determine absolute bias ( following e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) we first calculate the two - point autocorrelation of dark matter as a function of redshift .
we use the halofit code of @xcite to determine the nonlinear - dimensionless power spectrum @xmath159 of the dark matter assuming our standard cosmology , and the slope of the initial fluctuation power spectrum , @xmath160 .
the fourier transform of the @xmath159 gives us the real - space correlation function @xmath63 , which we then integrate to @xmath161 @xmath14 mpc following equation ( [ eqnwpint ] ) to obtain the dark matter projected correlation function @xmath162 .
the uncertainty in the dm power spectrum obtained from halofit is @xmath163% ; this corresponds to a systematic uncertainty @xmath70.05 dex in @xmath17 , but does not impact the relative halo masses of the different subsamples . to derive quasar absolute bias from the projected real - space correlation function ,
we average the @xmath164 over the redshift distribution of the sample , weighted by the overlap with the galaxy pdfs .
the overlap of each quasar with the galaxy pdfs is given by @xmath165 and the corresponding @xmath3 for the dark matter is given by @xmath166 where @xmath167 is the quasar redshift .
the redshift distributions for the qso-1s and obs - qsos are essentially identical ( the resulting @xmath168 values for the two samples differ by @xmath452% on all scales ) so for simplicity we use the same @xmath164 ( defined for the qso-1s ) for both sets of quasars .
we obtain the bias by calculating the average ratio between the best - fit power - law model and @xmath164 over the range of scales of 110 @xmath14 mpc , for which @xmath164 corresponds closely to a power law and is dominated by the two - halo term .
the observed clustering amplitude relative to the dark matter corresponds to @xmath169 , where @xmath170 and @xmath171 are the absolute linear biases of the quasars and sdwfs galaxies , respectively . to measure @xmath171 from the galaxy autocorrelation function ( or @xmath169 from the quasar - galaxy angular cross - correlation , described in [ qsoang ] ) , we require an estimate of the corresponding @xmath108 of the dark matter . to obtain @xmath108
we use limber s equation to project the power spectrum @xmath172 into the angular correlation @xcite .
specifically , we perform a monte carlo integration of equation ( a6 ) of @xcite to obtain @xmath108 for the dark matter .
the key parameter in this equation is @xmath173 where @xmath174 is the redshift distribution of the galaxies .
we calculate @xmath174 from the sum of the pdfs of the galaxies for which we perform the autocorrelation . in deriving the dark matter @xmath108 for the quasar - galaxy cross - correlation
, we replace @xmath175 with @xmath176 where @xmath177 is the distribution of quasar redshifts . for each angular correlation analysis
we compute the average ratio between the best - fit power law model and the dark matter @xmath108 on scales 110 , where @xmath108 is dominated by the two - halo term .
this ratio yields @xmath178 for galaxy autocorrelations or @xmath179 for quasar - galaxy cross - correlations . finally , we use @xmath170 and @xmath171 to estimate the characteristic mass of the dark matter halos hosting each subset of galaxies or quasars . @xcite
derive a relation between dark - matter halo mass and large - scale bias that agrees well with the results of cosmological simulations .
we use eqn .
( 8) of @xcite to convert @xmath180 to @xmath17 for the mean redshift of each subset of objects .
if we use a different relation between @xmath181 and @xmath17 @xcite , we obtain estimates for @xmath17 that are similar , although slightly larger by 0.20.3 dex ; these differences do not significantly affect our conclusions .
we note that to estimate @xmath17 , we have performed fits to the observed @xmath3 on scales of 0.312 @xmath14 mpc . in principle the dark matter and galaxy correlation functions can have somewhat different shapes such that the bias depends on the range of scales considered . if we limit the fits on scales 112 @xmath14 mpc , the results change by @xmath1075% , but with slightly larger uncertainties .
we also note that our estimates of @xmath17 are relatively insensitive to our choice of @xmath182 . if we change @xmath182 from 0.84 to 0.8 ( as favored by the more recent wmap cosmology , e.g. * ? ? ?
* ) our @xmath17 estimates for quasars and galaxies increase by @xmath183 dex .
in this section we discuss the results of the correlation analysis and the characteristic dark matter halo masses for galaxies and quasars .
we first calculate the cross - correlation of the full qso-1 and obs - qso samples with sdwfs galaxies .
the resulting @xmath3 values and best - fit models are shown in figure [ fcorr ] , and fit parameters are given in table [ tabcorr ] .
for both sets of the quasars the observed real - space projected cross - correlation is highly significant on all scales from 0.112 @xmath14 mpc , and the power law fits return @xmath184 , similar to many previous correlation function measurements for quasars ( e.g. , * ? ? ?
* ; * ? ? ?
* ) and galaxies ( e.g. , * ? ? ?
* ; * ? ? ?
the best - fit parameters are @xmath185 @xmath14 mpc , @xmath186 for the qso-1s , and @xmath187 @xmath14 mpc , @xmath188 for the obs - qsos .
the results indicate that the cross - correlation of the obs - qsos with galaxies is somewhat stronger than that for the qso-1s .
the corresponding values of @xmath179 are given in table [ tabcorr ] .
as a check , we also perform power law fits to @xmath3 but leaving the slope fixed to @xmath189 , which corresponds to the slope of the @xmath3 for the dark matter .
this also yields a significant difference in the clustering amplitude , although somewhat smaller , with @xmath190 and @xmath191 @xmath14 mpc for the qso-1s and obs - qsos , respectively .
( note that the formal uncertainties in @xmath148 here are smaller than for the above results because they do not account for covariance with @xmath149 . ) to obtain the absolute bias of sdwfs galaxies ( @xmath171 ) in order to extract the quasar bias @xmath170 from the cross - correlation results , we next derive the autocorrelation of sdwfs galaxies for the sample described in [ galauto ] .
the observed @xmath108 is shown in figure [ fgalang ] , along with the correlation function for dark matter , calculated as discussed in [ absbias ] .
fit parameters are given in table [ tabcorr ] .
the power - law model fits well on the chosen scales of 112 , although there is a clear excess corresponding to the one - halo term at @xmath192 , as is common in galaxy autocorrelation measurements ( e.g. , * ? ? ?
* ; * ? ? ?
the best - fit power law parameters are @xmath193 and @xmath194 , and the ratio of the best - fit power law to the dark matter @xmath108 yields @xmath195 or @xmath196 .
this accurate value for @xmath171 allows us to estimate @xmath170 for both types of quasars , based on the cross - correlation measurements .
we obtain @xmath197 and @xmath198 , for qso-1s and obs - qsos , respectively . converting these to dark matter halo masses using the prescription of @xcite as described in
[ absbias ] , we arrive at @xmath199)}= 12.7^{+0.4}_{-0.6}$ ] and @xmath5 for qso-1s and obs - qsos , respectively .
the difference is marginally significant ( @xmath200 , although as we discuss below , the obs - qso clustering may represent only a robust lower limit ) . for direct comparison with other studies that directly measure the quasar autocorrelation
, it is useful to present the quasar clustering in terms of effective power law parameters for their autocorrelation . assuming linear bias , the quasar autocorrelation can be inferred from the cross - correlation by @xmath201 ( e.g. , * ? ? ?
we can therefore use the power law fits to the quasar - galaxy cross - correlation and galaxy autocorrelation to derive an effective @xmath148 and @xmath149 for the quasar autocorrelation .
this yields @xmath202 @xmath14 mpc and @xmath203 for the qso-1s and @xmath204 @xmath14 mpc and @xmath205 for the obs - qsos .
the autocorrelation amplitude and @xmath17 for qso-1s are in excellent agreement with previous estimates for unobscured quasars , while the best - fit amplitude for obs - qsos is higher than most previous measurements of quasar clustering .
we compare these results to previous work and discuss possible interpretations in [ discussion ] .
lcccccccc + qso-1 & 445 & 1.27 & @xmath206 & @xmath207 & 1.1 & @xmath208 & @xmath209 & @xmath210 + obs - qso & 361 & 1.24 & @xmath211 & @xmath212 & 1.2 & @xmath213 & @xmath214 & @xmath215 + + qso-1 & 445 & 1.27 & @xmath216 & 1.8 & 1.1 & @xmath217 & @xmath218 & @xmath219 + qso-1 ( photo-@xmath6 ) & 445 & 1.27 & @xmath220 & 1.8 & 1.2 & @xmath221 & @xmath222 & @xmath223 + obs - qso & 361 & 1.24 & @xmath224 & 1.8 & 1.2 & @xmath225 & @xmath226 & @xmath227 + + galaxies & 151256 & 1.10 & @xmath228 & @xmath229 & 1.1 & @xmath230 & @xmath229 & @xmath231 + qso-1 & 445 & 1.27 & @xmath232 & 1.8 & 1.1 & @xmath233 & @xmath234 & @xmath235 + obs - qso & 361 & 1.24 & @xmath236 & 1.8 & 1.2 & @xmath237 & @xmath238 & @xmath239 anumber of objects include in the correlation analysis . for quasar - galaxy cross - correlation
, we use the full sample of 256,124 galaxies ( for @xmath3 calculations ) or 151,256 galaxies ( for @xmath108 calculations ) .
bmedian redshift for the objects included in the correlation analysis .
cuncertainties in the dm power spectrum introduce an additional systematic error of @xmath75% in @xmath180 ( and corresponding @xmath70.05 dex in @xmath17 ) .
further systematic errors in @xmath17 of @xmath70.2 dex are caused by uncertainty in @xmath182 and in the conversion from @xmath180 to @xmath240 , as discussed in
[ absbias ] .
however , these do not significantly effect the _ relative _ halo masses , so these uncertainties is not included here .
note that for fits with fixed @xmath149 , uncertainties on @xmath148 , bias , and @xmath17 do not account for covariance with @xmath149 and thus somewhat underestimate the error on the clustering amplitude .
dreal - space projected cross - correlation between quasars and galaxies , calculated as described in [ corranal ] .
for all @xmath3 calculations , error estimates for @xmath148 , bias , and @xmath240 include a 10% systematic uncertainty on the amplitude as described in [ crosscorr ] .
eangular galaxy autocorrelation and quasar - galaxy cross - correlation , calculated as described in [ corranal ] .
in this section we perform several tests to verify the validity of the clustering analysis outlined in [ corranal ] .
we first calculate the quasar - galaxy cross - correlation using a simple angular correlation function , minimizing dependence on the photometric redshifts .
we then check for variation in the observed clustering over the redshift range of interest and confirm that any variation is relatively weak .
finally , we estimate the effects of uncertainties on the photometric redshifts on the observed real - space clustering amplitude for the obs - qsos .
these checks confirm that our projected correlation analysis provides a robust estimate of the quasar - galaxy cross - correlation .
we first calculate the cross - correlation of quasars and sdwfs galaxies using a simple angular clustering analysis , and check whether the corresponding absolute bias is consistent with that derived from the more sophisticated @xmath3 calculation . to calculate the @xmath108 we use an estimator corresponding to equation ( [ eqnxidef1a ] ) but for cross - correlations : @xmath241 where each term is scaled according to the total numbers of galaxies and randoms . to maximize the signal - to - noise ratio by cross - correlating objects associated in redshift space ,
the galaxies include only the redshift - matched sdwfs sample of 151,256 objects described in [ galauto ] .
uncertainties are estimated using bootstrap resampling as described in [ uncertainties ] .
we fit the observed cross - correlation with a a power law as described in [ plawang ] . owing to the limited statistics which provide only very weak constraints on the power law slope , we fix @xmath242 ( corresponding to real - space @xmath243 ) .
the resulting cross - correlations and scaled dark matter fits are shown in figure [ fqsoang ] , and fit parameters are given in table [ tabcorr ] .
the estimates of @xmath148 and @xmath179 are in broad agreement between the two estimators , although as may be expected , the statistical uncertainties for the angular correlation analysis are larger ( by @xmath750% ) than for the real - space analysis with fixed @xmath149 . given that the absolute bias derived from the projected correlation function corresponds broadly to the bias from the noisier , but simpler angular cross - correlation , we conclude that there are no significant systematic effects that skew our estimate of @xmath3 .
our calculation of the real - space quasar - galaxy correlation function over the redshift range @xmath0 requires that @xmath3 varies only slowly between these redshifts , as discussed in [ crosscorr ] .
if the objects reside in similar halos at all redshifts , then we may expect @xmath148 to change slowly ; simulations suggest that the typical @xmath148 for the autocorrelation of dm halos of mass @xmath7@xmath18@xmath19 @xmath14 @xmath15 should change by @xmath244 @xmath14 mpc between @xmath245 and 2 ( see figure 10 of * ? ? ?
* ) . to test explicitly the redshift variation for the clustering in our sample , we re - derived @xmath3 using the method outlined in [ corranal ] but selecting quasars over smaller redshift bins of @xmath246 and @xmath247
uncertainties are calculated using the bootstrap method as for the full quasar samples , and dark matter and power - law fits are again performed over the range of separations @xmath248
@xmath14 mpc . we evaluate @xmath249 as in [ absbias ] , but only including the quasars in the redshift ranges of interest . owing to larger statistical errors and for simple comparison to the results over the full redshift range , in the power law
fits we fix @xmath149 to 1.8 .
the resulting @xmath3 for the separate redshift bins and the power law fits are shown in figure [ fwpz ] .
for the qso-1s we obtain @xmath250 @xmath14 mpc and @xmath251 @xmath14 mpc for the low- and high - redshift bins , respectively , and for the obs - qsos we correspondingly obtain @xmath252 @xmath14 mpc and @xmath253 @xmath14 mpc .
the measured quasar - galaxy cross - correlation should be largely independent between the two redshift bins .
although the quasars are cross - correlated against the same galaxy sample in each bin , the galaxy samples will be weighted toward higher and lower redshifts in the high- and low-@xmath6 bins , respectively .
for the high and low redshift bins , the best - fit @xmath148 values bracket those for the full redshift samples , and are broadly consistent within the uncertainties .
interestingly , the best - fit clustering amplitude for the obs - qsos increases with redshift while it decreases for the obs - qsos ; however , given the uncertainties we decline to speculate on any possible difference in redshift evolution between the two subsets .
overall , the results in the different redshift bins confirm that any variation in the observed @xmath3 is sufficiently weak over the redshift range of interest , so that the method outlined in [ crosscorr ] should provide a reasonable estimate of the average clustering amplitude over the full redshift range .
the primary uncertainty in our estimate of @xmath3 for the obs - qsos is the lack of accurate ( that is , spectroscopic ) redshifts and difficulty in estimating the photo-@xmath6 uncertainties from the neural net calculations . as described in [ corranal ] , in calculating @xmath3 for the obs - qso galaxy cross - correlation , we simply assume that obs - qsos lie exactly at the best redshifts output by the neural net estimator .
any uncertainties in the photo-@xmath6s or systematic offsets from the true redshifts could therefore affect the resulting clustering measurement .
the fact that we obtain very similar estimates of @xmath170 for the obs - qsos from a simple angular cross - correlation analysis as from our @xmath3 calculation suggests that uncertainties on individual photo-@xmath6s do not strongly affect our estimates of the quasar bias , as long as the overall distribution in redshifts for the obs - qsos is correct .
however , it is possible that very large discrepancies from the true photo-@xmath6s , or any systematic shift in the redshift distribution , could affect both the estimate of @xmath3 and the real - space clustering parameters derived from @xmath108 . to precisely explore the effect of these errors , we take advantage of the fact that we have an equivalent sample of objects ( the qso-1s ) that have a similar redshift distribution and for which the redshifts are known precisely from spectroscopy .
we can therefore adjust the redshifts of the qso-1s and re - calculate @xmath3 to determine how uncertainties or systematic shifts affect the observed correlation amplitude . as a simple first test
, we calculate the @xmath3 for the qso-1s using the photo-@xmath6s ( as shown in figure [ fspz ] ) rather than spectroscopic redshifts .
figure [ fagn5pz ] shows that the resulting @xmath3 differs little from that obtained using spectroscopic redshifts ; the clustering amplitude for a power - law fit with fixed @xmath189 is lower by 12% .
note that if we allow @xmath149 to float , the average bias for the photo-@xmath6 sample is actually larger by @xmath210% ( owing to a slightly flatter slope ) but well within the statistical uncertainties .
we conclude that for the qso-1s , photo-@xmath6 errors do not have a significant effect on our measurements of the clustering .
to explore photo-@xmath6 errors in more detail , we systematically test the effects of gaussian random errors in the quasar redshifts . for each quasar
, we shift the best estimates of redshift ( spec-@xmath6s for qso-1s and photo-@xmath6s for obs - qsos ) by offsets @xmath254 selected from a gaussian random distribution with dispersion @xmath255 . using these new redshifts we recalculate @xmath3 , using the full formalism described in [ corranal ] .
we perform the calculation ten times for each of several values of @xmath255 from 0.02 up to 0.2 ( which smears the redshifts across most of the redshift range of interest ) . to ensure that this step does not artificially smear out the redshift distribution beyond the range probed by the galaxies , we require that the random redshifts lie between @xmath0 ; any random redshift that lies outside this range
is discarded and a new redshift is selected from the random distribution . for each trial
we obtain the relative bias by calculating the mean ratio of @xmath3 , on scales 110 @xmath14 mpc , relative to the @xmath3 for the best estimates of redshift .
we then average the ten trials at each @xmath256 to obtain a relation between relative bias and @xmath256 , shown in figure [ ftest_fake_box](a ) .
as may be expected , figure [ ftest_fake_box](_a _ ) shows that shifting the qso-1 redshifts from their true values causes a decrease in the cross - correlation amplitude , as the quasars are preferentially correlated with galaxies that are not actually associated in redshift space .
we find a monotonic decrease in relative bias with @xmath256 , from @xmath257 for @xmath258 to @xmath259 for @xmath260 . repeating this calculation for
the obs - qsos reveals a very similar trend .
the decrease in bias with @xmath255 shown in figure [ ftest_fake_box](_a _ ) indicates that such errors would affect the measurements of the clustering amplitude by at most @xmath720% . while the above analysis suggests that random errors in the obs - qso photo-@xmath6s do not strongly affect the observed clustering amplitude , it is also possible that systematic uncertainties in the photo-@xmath6 ( consistent over- or under - estimates of the redshift ) could significantly alter the observed bias . to test this , we shift the redshifts of the quasars as discussed in [ randomerrors ] , but in place of random shifts , we compress all redshifts toward one end or the other of the @xmath0 range .
( this procedure allows us to test the effects of systematic shifts in redshift while keeping the same overall redshift range . )
the shift in redshift is defined by a redshift scaling parameter @xmath261 , such that @xmath262 as an additional check we also perform a simple linear offset @xmath263 of the redshifts , allowing the redshifts to move outside the selection range of @xmath0 . as in [ randomerrors ] , we use these new redshifts to recalculate @xmath3 via the full formalism described in [ corranal ] , and determine the relative bias on scales 110 @xmath14 mpc .
relative bias versus @xmath261 and @xmath264 are shown in figures [ ftest_fake_box](b ) and ( c ) .
for the qso-1s , the peak of the observed clustering amplitude is very close to @xmath265 , while shifting the redshifts down or up systematically decreases the bias .
the obs - qsos show a similar peak near @xmath265 , indicating that the obs - qso photo-@xmath6s are not systematically offset higher or lower than the true redshifts by a large factor .
we note that for the obs - qsos a slight shift to higher redshifts ( @xmath104 ) increases the clustering by a small amount ( @xmath28% ) . finally ,
we emphasize that any possible low - redshift contaminants , will serve only to decrease the observed clustering signal , as they will be completely uncorrelated in angular space with the higher - redshift sdwfs galaxies that lie in entirely separate large - scale structures . therefore the observed @xmath3 represents a robust lower limit to the clustering amplitude for the obs - qsos .
we have used the ir - selected quasar sample of to measure the clustering amplitude and to estimate characteristic dark matter halo masses for roughly equivalent samples of unobscured and obscured quasars .
we obtain highly significant detections of the clustering for both samples , with marginally stronger clustering for the obs - qsos . in this section
, we compare our results for qso-1s to previous results on unobscured quasar clustering , speculate on physical explanations for possible stronger clustering for obs - qsos , and discuss prospects for future studies with the next generation of observatories .
we compare our observed absolute bias for the mid - ir selected quasars to the clustering of optically - selected ( type 1 ) quasars , which has been well established by a number of works . among the most precise measurements to date are studies that have used data from the 2df and sdss surveys .
recently , @xcite measured the evolution of the quasar 3-d autocorrelation function based on spectroscopic quasars from sdss data release 5 , and compared to previous results from spectroscopic samples from the 2qz @xcite and 2slaq @xcite , as well as the clustering of photometrically - selected quasars from sdss @xcite .
figure [ fcompare](a ) shows the redshift evolution of linear bias for type 1 quasars from these studies , taken from figure 12 of @xcite .
where appropriate , we have converted the bias to our adopted cosmology using the formalism in the appendix of @xcite , assuming a shape factor @xmath266 for simplicity .
figure [ fcompare](b ) shows the corresponding estimates of characteristic @xmath17 derived from the linear bias using the prescription of @xcite .
the linear bias and halo masses for the qso-1s and obs - qsos are shown for comparison .
it is readily apparent from figure [ fcompare](a ) that the observed bias of type 1 quasars increases with redshift , as discussed in [ intro ] .
( these results are also consistent with a number of other quasar clustering studies using other data see e.g. , figure 15 of @xcite . )
the dashed curves in figure [ fcompare](a ) show the increase in bias with redshift for halos of constant mass , clearly showing that at all redshifts the qso-1s reside in dark matter halos of roughly a few @xmath267 @xmath14 @xmath15 .
our measurement for the qso-1s is in excellent agreement with the evolution in linear bias and roughly constant @xmath17 observed in previous measurements of type 1 quasar clustering . for the obs - qsos ,
the best - fit bias is marginally larger ( @xmath268 ) , corresponding to a factor of roughly four difference in the characteristic @xmath17 between the qso-1s and obs - qsos ( figure [ fcompare]b ) .
as discussed above , random errors in the photo-@xmath6s can only decrease the observed clustering amplitude and the inferred @xmath17 .
thus our measurement of obs - qso clustering represents a lower limit , and it is possible that the true obs - qso bias is somewhat higher ( although the results of [ verification ] suggest the true bias may be larger by at most @xmath720% ) . based on this analysis we can make the robust conclusion that the obs - qsos are _ at least _ as strongly clustered as their qso-1 counterparts . stronger clustering for obscured quasars would have significant implications for physical models of the obscured quasar population . in terms of unified models ,
a difference in clustering between obscured and unobscured quasars would rule out the simplest picture in which obscuration is purely an orientation effect , but may be consistent with more complicated scenarios where the effective covering fraction changes with environment .
alternatively , if obscuration is caused by large ( @xmath7kpc scale ) structures , then the processes that drive these asymmetries ( e.g. , mergers , disk instabilities , accretion of cold gas from the surrounding halo ) may be more common in halos of larger mass .
indeed , given that some fraction of quasars might naturally be expected to be obscured by a `` unified''-model torus , any observed differences in clustering may reflect even stronger intrinsic dependence of large - scale obscuration and environment .
an intriguing scenario for obscured quasars is that they represent an early evolutionary phase of rapid black hole growth before a `` blowout '' of the obscuring material from the central regions of the galaxy and the emergence of an unobscured quasar ( e.g. , figure 1 of * ? ? ?
quasars tend to radiate at large fraction of the eddington rate ( @xcite ; although see @xcite ) , so that the similar @xmath50 for qso-1s and obs - qsos would imply that they host black holes of similar masses .
any correlation ( e.g. , * ? ? ?
* ; * ? ? ?
* ) , even if indirect ( e.g. , * ? ? ?
* ) , between the final masses of black holes and those of their host halos would thus suggest that our obscured and unobscured quasars would have the same @xmath17 , as long as their black holes are near their final masses .
however , if obscured quasars are in an earlier phase of rapid growth and so are in the process of `` catching up '' to their final mass ( e.g. , * ? ? ?
* ) , then they would have a larger @xmath17 compared to unobscured quasars with the same @xmath269 . in light of recent debate as to whether black holes generally grow before or after their hosts ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , this scenario would imply that black hole growth lags behind that of the host halo . in any physical picture ,
a significant difference in clustering between obscured and unobscured quasars would also imply a difference in accretion duty cycles ( or equivalently , lifetime ) . qso-1s and obs - qsos
are found in roughly equal numbers , but the abundance of dark matter halos drops rapidly with mass ( e.g. , * ? ? ?
* ) , thus implying that if one type of quasars are found in larger halos then they must be longer - lived . with our current results , we are able to rule out any model in which obscured quasars are substantially _ less _ strongly clustered or have _ shorter _ lifetimes than their unobscured counterparts . with more accurate future measurements , detailed studies of halo masses and lifetimes for obscured and unobscured
quasars could place powerful constraints on evolutionary scenarios such as those described above .
our results demonstrate the potential for studying the clustering of obscured quasars in extragalactic multiwavelength surveys , and the marginally significant difference in clustering we observe for obscured and unobscured quasars provides strong motivation for more precise measurements in the future .
the two main avenues for progress are improvements in redshift accuracy and selection of larger samples for better statistical accuracy .
upcoming sensitive , wide - field multi - object spectrographs will enable efficient measurements of redshift for large numbers of optically - faint sources and so improve calibrations of obscured quasar photo-@xmath6s , or with large enough samples , enable fully 3-d clustering studies .
in addition , we will soon have the capability to detect many thousands of obscured quasars based on very wide - field observations in the mid - ir with the _ wide - field infrared survey explorer _ @xcite and in x - rays with erosita @xcite or the _ wide - field x - ray telescope _ @xcite . these data sets will allow us to measure obscured quasar clustering with statistical precision that is comparable to current measurements of unobscured quasars .
we have used data from the botes wide - field multiwavelength survey to measure the two - point spatial cross - correlation between unobscured ( qso-1 ) and obscured ( obs - qso ) mid - ir selected quasars in the redshift range @xmath0 . the qso-1s exhibit clustering corresponding to a typical @xmath270 @xmath14 @xmath15 , similar to previous studies of optically - selected quasar clustering .
we robustly determine that the obs - qsos are clustered _ at least _ as strongly as the qso-1s , with a marginally stronger signal corresponding to host halos of mass @xmath271 @xmath14 @xmath15 ; the true clustering amplitude could be up to @xmath720% larger owing to photo-@xmath6 uncertainties for the obs - qsos that can decrease the observed correlation amplitude .
our results motivate more accurate measurements of obscured quasar clustering with larger quasar samples and more accurate redshifts .
if future studies confirm that obscured quasars are more strongly clustered than their their unobscured counterparts , this would rule out the simplest `` unified '' models and may provide evidence for scenarios in which rapid obscured accretion represents an evolutionary phase in the growth of galaxies and their central black holes .
we thank our colleagues on the ndwfs , ages , sdwfs , and xbotes teams .
we thank the anonymous referee for helpful comments that improved the paper , and philip hopkins and peder norberg for productive discussions .
the noao deep wide - field survey , and the research of a.d . and b.t.j . are supported by noao , which is operated by the association of universities for research in astronomy ( aura ) , inc . under a cooperative agreement with the national science foundation .
this paper would not have been possible without the efforts of the _ chandra _ , _ spitzer _ , kpno , and mmt support staff .
optical spectroscopy discussed in this paper was obtained at the mmt observatory , a joint facility of the smithsonian institution and the university of arizona .
the first _ spitzer _
mips survey of the botes region was obtained using gto time provided by the _
spitzer _ infrared spectrograph team ( pi : james houck ) and by m. rieke .
we thank the collaborators in that work for access to the 24 micron catalog generated from those data .
r.c.h . was supported by an stfc postdoctoral fellowship and an sao postdoctoral fellowship , and a.d.m .
was generously funded by the nasa adap program under grant nnx08aj28 g . d.m.a .
is grateful to the royal society and philip leverhulme prize for their generous support
. r.j.a . was supported by the nasa postdoctoral program , administered by oak ridge associated universities through a contract with nasa .
the large area of the botes survey allows us to measure galaxies at relatively large physical separations ; 1 degree corresponds to 50 @xmath14 mpc at @xmath272 .
however , when we use equation ( [ eqnwp ] ) to calculate the projected real - space correlation function on large scales , we find that the @xmath3 flattens on scales @xmath27320 @xmath14 mpc , corresponding to tens of arcmin , and then becomes negative at @xmath274 @xmath14 mpc ( figure [ fwprand ] ) . this behavior is not observed for quasar clustering from other , wider - field surveys ( e.g. * ? ? ?
* ; * ? ? ?
* ) , for which the correlation function continues to decrease on larger scales .
while the integral constraint require that the correlation function becomes negative on some scales , in galaxy auto - correlation surveys this generally only happens at @xmath275 200 @xmath14 mpc ( e.g. , * ? ? ?
one possibility is that the observed behavior is due edge effects arising from the finite geometry of the sdwfs field , which are not taken into account by the simple @xmath276 estimator in the formalism ( e.g. , * ? ? ?
* ) . to test this possibility ,
we re - performed the correlation analysis described in
[ corranal ] , after randomizing the positions of the quasars on the sky within the area of good sdwfs photometry .
we performed 10 separate random trials , for which the cross - correlations are shown in figure [ fwprand ] along with the @xmath3 values for the qso-1s and obs - qsos .
it is clear from figure [ fwprand ] that on scales @xmath10710 @xmath14 mpc , the projected cross - correlation between the random quasars and galaxies is small compared to the @xmath3 for the real quasar sample .
however , on scales @xmath27720 @xmath14 mpc , both the real and random samples show an increase in @xmath3 which eventually becomes negative around @xmath278 @xmath14 mpc .
the quantities of interest in this paper ( i.e. absolute bias and dark matter halo mass ) can be measured by studying the correlations on scales @xmath4512 @xmath14 mpc , where the artifacts are small and have negligible impact on the fits to the correlation function . for this paper
we therefore limit the correlation analyses to those scales . , b. t. & dey , a. 1999 , in asp conf .
ser . 191 : photometric redshifts and the detection of high redshift galaxies , ed .
r. weymann , l. storrie - lombardi , m. sawicki , & r. brunner ( san francisco : asp ) , 111 | we present the first measurement of the spatial clustering of mid - infrared selected obscured and unobscured quasars , using a sample in the redshift range @xmath0 selected from the 9 deg@xmath1 botes multiwavelength survey . recently
the _ spitzer space telescope _ and x - ray observations have revealed large populations of obscured quasars that have been inferred from models of the x - ray background and supermassive black hole evolution . to date , little is known about obscured quasar clustering , which allows us to measure the masses of their host dark matter halos and explore their role in the cosmic evolution of black holes and galaxies . in this study
we use a sample of 806 mid - infrared selected quasars and @xmath2250,000 galaxies to calculate the projected quasar - galaxy cross - correlation function @xmath3 .
the observed clustering yields characteristic dark matter halo masses of @xmath4 ) = 12.7^{+0.4}_{-0.6}$ ] and @xmath5 for unobscured quasars ( qso-1s ) and obscured quasars ( obs - qsos ) , respectively .
the results for qso-1s are in excellent agreement with previous measurements for optically - selected quasars , while we conclude that the obs - qsos are _ at least _ as strongly clustered as the qso-1s .
we test for the effects of photometric redshift errors on the optically - faint obs - qsos , and find that our method yields a robust lower limit on the clustering ; photo-@xmath6 errors may cause us to underestimate the clustering amplitude of the obs - qsos by at most @xmath720% .
we compare our results to previous studies , and speculate on physical implications of stronger clustering for obscured quasars . |
infrared radiation from molecules and grains in space is usually or mostly attributed to heating by visible and uv light absorption .
while this may be justified by the ubiquity of light radiation , it must be remembered that the ism ( interstellar medium ) is also permeated with atomic hydrogen .
h atoms are also found in abundance in the limbs of molecular clouds illuminated by bright young stars ( pdrs or photodissociated regions ) .
this paper accordingly explores the relative importance of deposition of energy by atomic hydrogen impacts on the same targets .
as is well known , the formation of molecular hydrogen by two isolated h atoms is forbidden by energy and momentum conservation ; it is therefore associated instead with recombination in presence of a third body * ( see spitzer 1977 ) .
the most readily available such body is a hydrocarbon molecule or grain .
the latter is made of a carbon skeleton to which peripheral h atoms are attached .
an incident h radical , if moderately energetic , will attract one of the h atoms at the grain periphery forcefully enough to capture it and form an h@xmath0 molecule which readily escapes into space ; this is called h abstraction " in surface physics language .
it will be shown below that this process does not leave much energy in the grain .
however , it does leave an unoccupied dangling " c - bond .
when the next incident h atom meets that free bond , it is most likely to form a strong chemical bond with the host c atom .
this recombination " deposits in the grain nearly 4 ev , * half of which is available in the form of kinetic energy to excite the vibrations of the hydrocarbon particle ( the other half going into potential energy ) .
although this is no more energy than is carried by a visible photon , it is totally expendable in vibrational excitation , by contrast with photonic energy , which is first delivered to electrons , to relax thereafter into continuum and vibrational radiation .
the emission of vibrational bands following h - impact excitation is a type of chemiluminescence . * the basics of this process
were first put forward by guillois et al .
it may be considered as a special case of the radical reactions " previously envisioned by allamandola and norman @xcite ; but these authors did not treat any process in particular , dwelling instead on relaxation times and emission probabilities . * * * the present work is also motivated by several astronomical observations which hint at the need to complement the photon flux with another excitation agent for the emission of uibs ( unidentified infrared bands ) .
one is the fact that the relative intensities of the bands in the spectrum of uibs do not change notably with the average photon energy of the ambient radiation ( uchida et al .
1998 ) ; besides , neither the paucity of uv radiation ( as in galaxy m31 ; pagani et al . 1999 ) nor the absence of starbursts ( see haas et al .
2002 ) precludes uib emission .
another hint was the comparison , by onaka et al .
@xcite , using the irts satellite , of uib emission ( 5 to 12 @xmath1 m ) at various galactic latitudes throughout the galaxy : the intensity of emission decreases quickly with increasing latitude , but its spectral profile remains unchanged . while there are no hot stars at these latitudes ( and therefore much less far uv photons ) , the uib intensity follows the trend of h atom density , which extends farther from the galactic plane than does molecular hydrogen density ( see imamura and sofue 1997 ) .
earlier on , boulanger and prault @xcite , studying the solar environment by means of the iras satellite , found that , away from heating sources and molecular clouds , the ir emission from the cirruses of the ism is well correlated with the column density of hi gas .
pagani et al .
@xcite reached the same conclusion from the study of a large number of sight lines through the near - by andromeda nebula ; by contrast , they found no correlation of ir emission with uv flux .
still another remarkable fact is the occurrence of the uib intensity peak , in pdrs ( photo - dissociation regions ) seen edge - on , in between the peak of recombination radiation ( signaled by h@xmath2 lines , e.g. br @xmath3 ) , on the star - illuminated side ( hii region ) , and the peak of h@xmath0 de - excitation radiation at 2.42 @xmath1 m .
this is precisely where maximum atomic h density is to be found .
the generic example is the orion bar ( see roche et al .
1989 ) , sellgren et al .
1990 , graham et al .
1993 ) . * an idealized illustration of the succession of regions and of the distribution of densities and radiative fluxes was given by guillois et al .
@xcite . *
* finally , the need for some excitation process other than photon absorption to understand available observations seems to have been felt by other workers ; e.g. duley and williams @xcite reconsidered briefly another type of chemical excitation of ir vibration : the sudden and violent release , upon mild heating , of potential energy stored in molecules in the form of radicals . * in sec .
2 , i describe the elementary processes which create conditions under which h atom excitation becomes operational . in sec . 3 , rate equations coupling the relevant physical quantities are laid down and the statistical equilibrium values of the variables are determined .
the latter are used in sec . 4 to determine the energy build up in a grain , under h atom bombardment .
section 5 displays expressions for the uib reemission power . at this stage
, it becomes possible to compare uv and h impact excitation of grains , which is done in sec .
finally , sec . 7
compares predictions of the chemiluminescence model with observations in various environments and measurements of different physical quantities .
when an h atom ( radical ) impinges upon a hydrocarbon target , a number of processes may occur ( see lohmar et al .
2009 , cazaux and tielens 2002 , papoular 2005 ) : a)recoil ; this is usually nearly elastic , i.e. it involves little energy exchange .
b)recoil with expulsion of an h atom off the target ; this breaking of a c - h bond requires an unusually energetic h projectile ( @xmath430 000 k ) .
c)h abstraction , in which the incoming h atom is attracted by a target h atom to form a h@xmath0 molecule which readily leaves the target with a high velocity ; this leaves a dangling ( unoccupied ) c bond at the target surface , together with a fraction of the c - h bond energy .
d)surface recombination of the h projectile with a target dangling bond ( chemisorption ) ; this may occur over a large range of projectile energies , and leaves the whole c - h bond energy in the target .
\e ) sticking ( physisorption ) of the incoming h atom at the target surface , over which it can wander until it goes through ( a ) , ( c ) or ( d ) .
the steady state coverage of the target depends on all 5 processes but the energy balance is not the same for all , as shown below , * using version 7.5 of hyperchem , the chemical simulation code commercialized by hypercube , inc .
earlier versions of this code were used and described in detail by , for instance , papoular @xcite and @xcite .
here , i used the semi - empirical pm3 method of molecular dynamics calculation provided by the package .
this method uses quantum - mechanical calculations , supported where necessary by empirical constants from the chemistry laboratory , which are kept up to date . in pm3 ,
the empirical part was tailored by comparison with carbon - bearing molecules , so the energies given by this code agree with experiment to a few percent , better than necessary for present purposes . *
first , the c - h bond energy was determined in several different structures .
this was done by first optimizing the structure of interest ( at 0 k ) , i.e. minimizing its potential energy as a function of geometry , and measuring this energy .
the h atom was then extracted and the new , higher , potential energy measured .
the difference is the required bond energy , @xmath5 . for h@xmath0 ,
this is 109.4 kcal.mol@xmath6 . for an aliphatic chain ( -ch@xmath0-ch@xmath7 ) , @xmath5=98 ; in methyl ch@xmath8 , 101 to 106
; in benzene , 116 to 118 , all in kcal.mol@xmath6 ( 1 ev=23 kcal.mol@xmath6 ) . * in the following , we take @xmath5=4 ev , uniformly , as higher accuracy is not warranted in describing the general properties of the present model . consider first case ( d ) .
the target is assumed to be initially at rest at 0 k. in a simplified semi - classical picture , the representative point of this state is the bottom of a parabolic potential well * as in fig .
1.when the incoming h atom is captured by a dangling bond , the potential well is suddenly depressed by @xmath5 , so the representative point is now found above the new bottom by the same quantity , with zero ke ( kinetic energy ) .
it is compelled to slide down the well , rise on the other side , fall down again and repeat this sequence indefinitely . in this new , dynamic
, state the average ke and pe ( potential energy ) are both equal to @xmath9 .
this energy is initially localized in the new c - h bond in the form of large amplitude stretching vibrations , but it quickly spreads to other , closely coupled , vibrations of the structure * * the average life time of this state before energy is lost through ir radiation is usually longer than 1 ms * ( see radzig and smirnov 1980 ) .
so the posibility must be considered that , before this happens , another h radical is captured by the same target .
the representative potential well will again be depressed .
assume , for simplicity , that the transition to the new , third , state occurs at the moment when the representative point is at its peak height in the well with pe=@xmath5 and ke=0 .
its instantaneous pe above the bottom of the new well will therefore be @xmath10 , and the average pe and ke will rise to @xmath5 .
thus for each successive h recombination , the target particle gains @xmath11 ev in vibrational energy . if this energy were shared equally between , say n=100 atoms of the target , this would correspond to a * vibrational temperature of @xmath12170 k. * * consider now case ( c ) . here , the gain in pe of the target upon loosing an h atom is partially compensated by the loss of pe associated with the formation of a h@xmath0 molecule , because the h bond energies involved are almost equal .
moreover , the h@xmath0 molecule carries away vibrational , translational and rotational kinetic energies .
typical values for an aliphatic chain , for instance , are , respectively , 19 , 15 and 0.3 kcal / mol for the h@xmath0 molecule , leaving only a few kcal / mol in the target . for case ( b )
, one can develop the same argument as for ( d ) , except that , here , the target potential well is suddenly raised , instead of depressed , upon losing an h radical .
* assume , in fig .
1 , that the system is given more energy than what is necessary to break a ch bond .
after a while , half this extra gain in ( potential ) energy is converted into ke , as in case ( d ) . however
, this process clearly requires an unusually energetic h projectile . * in estimating the overall vibrational energy gain , below
, we will therefore neglect the contributions of recoil and h absraction by comparison with that of h recombination .
by hi region is meant , here , a region of space permeated with photons energetic enough to dissociate molecular hydrogen without ionizing the atoms .
this excludes dense molecular clouds but includes , for instance , cirruses in the diffuse ism .
let @xmath13 be the number densities ( @xmath14 ) of atomic , molecular and nuclear hydrogen , respectively , with @xmath15 .
assume the average velocity of the h atoms is @xmath16 cm.s@xmath6 , and the flux of dissociating photons is @xmath17 @xmath18s@xmath6 .
let @xmath19 and @xmath20 be the number of c and h atoms in each hydrocarbon grain , and @xmath21 the number density of such grains ( @xmath14 ) . assume , for simplicity , that , in the ism , the mass ratio between grain and gas is 1/300 .
ignoring heteroatoms , which are quite rare in fact , and h atoms which are much lighter , this implies @xmath22 , or @xmath23 .
since we are interested in h recombination with naked c atoms ( those exhibiting a dangling bond ) at the grain surface , we have to distinguish between the naked and the h - covered " c atoms .
this is illustrated in fig .
2 , for an aromatic and an aliphatic skeleton . roughly speaking
, a covered c atom may anchor one or two h atoms according to whether the skeleton is aromatic or aliphatic ; so there is 0.5 to 1 covered c atom for every bound h atom . taking 0.75 for simplicity , and assuming the carbon skeleton is not too compact ,
we may therefore write @xmath24 on average . now , the relative numbers of covered and naked c atoms depend on the density of gaseous h atoms above the grain , and on the various c - s ( cross - sections ) of the gas - grain reactions . in statistical equilibrium ,
the number of gaseous h atoms captured by a grain is equal to the number of bound h atoms liberated from the grain by h impact .
this can be written @xmath25 where @xmath26 are c - s s for recombination , abstraction and expulsion , respectively ; @xmath27 is the c - s for recombination after sticking ( indirect recombination " ) ; a factor @xmath28 was omitted from each side as it is the same for all reactions , so this equilibrium is independent of the ambient gas density .
then , defining @xmath29 , we get @xmath30 and , hence , @xmath31 .
the c - s s * involved in gas - grain interactions have been estimated , computed , modelled or measured ( see cazaux and tielens 2002 , * mennella et al .
2002 , papoular 2005 , lohmar et al .
the results concur only approximately . here
, we read from papoular @xcite , with @xmath32 cm.s@xmath6 : @xmath33 @xmath34 , @xmath35 @xmath34 and @xmath36 .
equation 2 then gives @xmath37 , indicating that the hydrogen coverage of the target is @xmath38 .
* by way of comparison , in a study of gaseous atomic hydrogen interactions with carbon grains , papoular @xcite found that the h coverage of hydrocarbon grains varies between @xmath39 and @xmath40 , depending mainly on the kinetic temperature of the h gas . in a recent study of the hydrogenation of hac ( hydrogenated amorphous carbon ) duley and williams
@xcite estimated the coverage at 0.3 - 0.5 . * * * turning to the gas densities , let @xmath41 be the h@xmath0 molecule c - s for dissociation by photons .
the rate equations for gaseous atomic and molecular hydrogen densities can be written @xmath42 the terms in the r.h.s .
of the first of these equations are for photodissociation , abstraction , recombination and expulsion respectively ; in the second line , the terms are for photodissociation and abstraction respectively . * in statistical equilibrium , the derivatives in the l.h.s . must be set to 0 .
defining * @xmath43 , * and using the second rate equation , * @xmath44 * finally , using @xmath15 , we obtain * @xmath45 note that , while @xmath46 depends only on c - s s , @xmath47 also depends on the radiative flux and on the velocity of h radicals . as for the dissociating photon flux which enters the rate equations ,
its value * varies considerably with the wavelength in the spectral region of interest ( lyman continuum , 900 - 1100 ) , and so does the dissociation rate @xmath48 ( see draine 1978 , fig . 3 , and references therein ) .
it is therefore usual to replace the latter by an estimated average value , @xmath49 , in this wavelength range .
we adopt @xmath50 s@xmath6 for the galactic ism ( see stecher and williams 1979 , jura 1974 ) .
nearer to hot stars , @xmath51 , where @xmath52 is the galactic fuv ( far ultra - violet ) flux , and @xmath53 may reach up to @xmath54 .
we shall take @xmath55 , assuming that the * uv spectral profile does not change as the radiation varies in intensity .
using these parameter values , eq .
3 then becomes * * @xmath56 thus , _ the ratio of molecules to atoms , z , is proportional to the total number density of h nuclei , and inversely proportional to the uv flux_. this simplistic treatment neglects the extinction of dissociating light by molecules and grains along its path , as well as self - shielding of the h@xmath0 molecules against uv due to their excitation out of their ground level ( see draine and bertoldi 1996 , fig . 8) .
when these effects are included , it is found that , for given density and radiation flux , the light is almost unaffected up to a critical column density , beyond which it is quickly extinguished while the molecular fraction increases steeply .
the elaborate theory helps understanding the earlier observations of molecular cloud limbs ( see savage et al .
1977 , fig .
6 ) , which set the critical column density at a few times @xmath57 @xmath18 .
the fortunate conclusion for our present purposes is that our treatment is acceptable below the critical column density , i.e. in most environments from which the uibs are usually observed to originate : cirruses , molecular limbs and pdrs .
assuming the frequency of collisions between grains to be much lower than the frequency of h impacts on a particle , the energy deposited by the latter accumulates in the particle until it is reemitted in the form of ir radiation , which is its sole outlet .
a fundamental quantity is , therefore , the total kinetic energy deposited per atom during the radiative lifetime . from eq .
2 , the total number of atoms per particle is @xmath58 from sec .
2 , the vibrational energy deposited by an impinging h atom that recombines on the particle is @xmath9 . the number of such events that occur before ir emission is proportional to the radiative lifetime , @xmath59 , to the ambient h atom influx , to the dangling bond coverage over the particle surface and to the c - s for direct and indirect recombination . on average , the final ( or maximum ) vibrational energy
_ per target atom _ will be @xmath60\,,\ ] ] or , * using eq . 2 and 3,@xmath61
* the characteristic behaviour of this function is determined by the factor ( @xmath62 ) . as @xmath47 tends to 0 ( high @xmath63 ) ,
@xmath64 which is proportional to @xmath65 but no longer depends on @xmath53 . as @xmath47 increases beyond 1 ( low @xmath63 ) , @xmath66 which is proportional to @xmath53 but
no longer depends on @xmath65 . for the parameters of our numerical example , and @xmath67 s , @xmath68 and @xmath69 , if @xmath65 is in @xmath14 and @xmath70 in ev
for a direct comparison with observations , the total uib emission per unit volume and time is needed . assume that all the chemical energy carried by the impinging h atoms is converted into ir emission .
but , for the grains to distinctly emit the uibs , they must be * neither too small ( in which case , they could not contain all the functional groups that carry the uibs ) , nor too large ( in which case most of the energy would go into the continuum ) : this is demonstrated by emittance mesurements on carbonaceous particles of different compositions and sizes ( see , for instance , solomon et al .
thus , only a fraction , @xmath71 , in mass , of the is carbonaceous grains ( as defined by these two constraints ) must be included in our calculation . using eq .
8 , the uib emission can then be written * @xmath72 since @xmath70 is proportional to @xmath59 ( see eq .
6 ) , @xmath73 no longer depends on the latter .
figure 3 displays @xmath74 , the uib power emitted per ambient h nucleus .
again , we distinguish regions of low and high illuminations and write @xmath75 @xmath76 where @xmath65 is in units of @xmath14 and @xmath73 is in w.m@xmath77 ( choices imposed by observational practice ) in a given is environment , it may be helpful to determine which of the two cases applies .
equation 8 clearly shows that this depends on whether @xmath47 is much larger or much smaller than 1/2 .
the critical condition , @xmath78 , can be expressed in terms of @xmath53 and @xmath65 , using eq . 5 :
@xmath79 which is displayed in fig .
it is apparent , from eq . 5 and 11 , and fig .
4 , that , in the local ism ( @xmath80 ) , dissociation is rather high and that eq .
9 and 12 are the better approximations .
it is instructive to determine the relative contributions of h atoms and dissociative uv photons to grain heating .
this is done here by considering the ratio of the rates of h atom recombinations ( on grains ) and uv photon absorptions per unit grain surface , * multiplied by the corresponding energies deposited in the grain , @xmath81 where @xmath82 is the fuv absorption cross - section per c atom . for c - h bonds , this is @xmath83 ( turro 2009 ) , while for c = c bonds it rises to @xmath84 @xmath34 ( see roche et al .
@xcite ) . for graphite
, it is @xmath85@xmath34 .
let us adopt @xmath86 .
also take , as above , @xmath9=2 ev , @xmath32 cm.s@xmath6 , @xmath35 @xmath34 and @xmath87=0.5 . the average energy of dissociating ( lyman continuum ) photons will be taken to be 12 ev , and @xmath88 photons .
@xmath18.s@xmath6 for the diffuse ism ( @xmath80 ; see draine 1978 and references therein ; mezger et al . 1982 ) .
this gives @xmath89 , or , using eq .
4 for @xmath90 , in the two extreme cases ( @xmath47 large or small ) : * @xmath91 @xmath92 the determining factor in eq .
15 is @xmath93 .
thus , _ in small fuv fluxes , indirect heating by h atom impacts is much more efficient than radiative heating , independent of @xmath53 ; this is essentially because there are so many fast hydrogen atoms that their flux is larger than that of fuv photons .
but , in high fuv fluxes , hydrogen is almost totally dissociated , so @xmath94 decreases because , as @xmath53 increases , @xmath90 no longer increases while radiative heating still increases and ultimately becomes dominant .
_ this conclusion holds even if some less energetic photons ( 1100 to 2400 @xmath95 ) are included ; it is even strengthened when it is recalled that photon absorption is intrinsically less efficient than h impact in producing uib emission , as some of its energy is diverted into ionization and another part into continuum emission .
the ratio ( 15 ) can be slightly modified to become the ratio of total uib emission to total ir emission . to that effect ,
assume , as in sec . 5
, that total conversion of the chemical energy into uib emission occurs only in a fraction @xmath71 , in mass , of the carbonaceous grains .
then , the numerator of eq .
15 must be multiplied by @xmath71 in order to represent total uib emission . on the other hand ,
note that not only dissociating photons ( 900 to 1100 @xmath96 ) , but also weaker uv photons ( 1100 to 2400 @xmath96 , according to the definition of the habing uv unit of flux ) , contribute to the radiative heating of grains , and must also be included in the photon flux , @xmath17 ; for the latter , we now take @xmath97 * ( based on fig . 3 of draine 1978 )
, so the denominator now nearly represents the total ir emission .
expressions 15 to 17 are thus multiplied by @xmath98 .
this will be compared with observations in sec
we seek now to compare the predictions of the model with relevant observations .
consider first the diffuse infrared emission " from the galaxy , as measured by boulanger and prault @xcite in the solar neighbourhood .
they compared the ir measurements of the iras satellite , in its 4 bands , 12 , 25 , 60 and 100 @xmath1 m , with the hi radio emission from the same regions of the solar neighbourhood , away from heating sources and outside dense molecular clouds , where @xmath80 by definition . noting that the radio emission is proportional to the h atoms number density , they deduced the band emission per h atom , in each band , and found 1.1 , 0.7 , 1.1 and 3.2 , respectively , in units of @xmath99 w per h atom , for a total of 6.1 .
boulanger and prault @xcite also derived an estimate of the emission per h atom of the diffuse ism , in the uib range ( 2 to 15 @xmath1 m ) : @xmath100 w.h atom@xmath6 . from these data
, @xmath71 can be deduced . for the contribution of uv photons to direct
grain excitation is negligible in the diffuse ism ( see sec .
6 ) , so both uib and total ir emissions are excited by h impact ( in the present model ) . as a consequence ,
their ratio is precisely equal to @xmath71 , according to our definition of the latter ( sec .
5 ) , so @xmath101 this considerable fraction is a necessary consequence of the assumption that uib and other ir emissions are excited by the same agent . this constraint can be met with h atom excitation because even relatively large grains can be excited ; the upper limit to their size is several thousands of c atoms per grain , before the continuum confiscates most of the chemical energy . in fact
, the grain efficiency in exciting uibs must be a decreasing function of grain size , so the transition between uib - emitting and non - uib - emitting must be gradual rather than a sharp cut - off .
pagani et al .
@xcite made a similar study of the diffuse ism of the andromeda nebula , m 31 .
they considered the signal from the lw2 ( 5 - 8 @xmath1 m ) filter of isocam on board the iso satellite as a tracer of uibs and found that _ it is extremely well correlated , across the galaxy , with the distribution of neutral gas as mapped by the hi ( 21 cm radio ) and co(1 - 0 ) signals ( @xmath65 in our model ) , which they also plotted .
by contrast , the correlation is poor with the ionized gas as seen through its h@xmath3 emission , and non - existent with the uv emission_. they conclude that the mid - ir emission can be excited by the ambient visible and near - ir radiation .
here , i argue that , if that were the case , there would still be a much tighter correlation with the young stars of the nebular ring than shown by the maps . instead , i set out to interpret quantitatively the findings of pagani et al .
@xcite by means of the present model .
pagani et al . analyzed their images , pixel by pixel , and plotted their measurements as a graph of i(lw2 ) as a function of the column density , n(h ) , in their fig .
6 . from this , they deduced the regression line @xmath102 * where @xmath103 , and i(lw2 ) is in mjy / pixel and @xmath104 , in @xmath18 .
given the size of the pixel ( 6``x6 '' ) , and neglecting the ordinate at the origin , -0.23 , we deduce @xmath105 * since the isocam images cover mainly the general ( diffuse ) ism of m 31 , this number can be compared with the value , 1.5@xmath106 , measured by boulanger and prault @xcite in our galaxy , in the uib range ( 2 to 15 @xmath1 m ) .
they are of the same order of magnitude , but the signal from m 31 is weaker by a factor @xmath1074 .
this must be due , in large part , to the fact that the lw2 filter covered the range 5 - 8 @xmath1 m , * thus excluding the strong 11.2 and 12.7 @xmath1 m bands , while the measurement of the galaxy covered the larger range 2 - 15 @xmath1 m .
the remaining difference is discussed next .
* figure 6 of pagani et al . exhibits a significant dispersion about the regression line .
this is quantified in their fig .
9 , which plots @xmath108 ( * our @xmath109 ) as a function of the fuv flux in a band of 15 nm width around 200 nm . when translated in the units used here
, their measurements extend over * the ranges 0.25 to 0.8@xmath110 erg.@[email protected]@xmath112 for the fuv flux , and 0.32 to 4.75@xmath113 w/ h atom , for @xmath109 .
* boulanger and prault @xcite , had also found considerable patchiness in the ir emission from the solar neighbourhood in all directions ( their fig .
this was associated with the notion of cirrus " imagined by low et al .
@xcite by analogy with this type of tenuous clouds in our atmosphere .
* * * in the present excitation model , this dispersion can be interpreted in terms of h density differences between regions having the same uv illumination . in order to * justify this proposition ,
we return to eq . 11 , which gives the exact expression of the power reemitted in the uibs per unit volume of space , insert 0.25 for @xmath71 and divide by 2 to take into account the smaller band width of the lw2 filter
. the result can be cast as * @xmath114 where @xmath115 , * as in eq . 5 , and @xmath65 is in @xmath14 .
* clearly , @xmath116 varies with @xmath65 and @xmath53 . as an example , take the minimum measured uv flux in fig .
9 * of pagani et al .
@xcite ( for which the dispersion of plotted points is greatest ) : 2.4 @xmath117 erg.s@[email protected]@xmath112 , which is close to the value in the solar neighbourhood : 2.1 10@xmath119 ( see draine 1978 ) . for lack of better information ,
assume that the ratio of dissociating uv flux to the flux at 200 nm is the same in m 31 as in the solar neighbourhood .
then , for the selected points , @xmath120 , according to our definition of @xmath53 in sec .
3 . for the corresponding minimum and maximum values of @xmath116 , eq . 19
gives , respectively , about 5 and 77 h at.@xmath14 . for the average @xmath116 ( eq .
20 ) , @xmath121 @xmath14 .
these calculations implicitely assume that every measured ir power emanates from a limited region of space ( a cirrus ) , with a relatively homogeneous density distinctly different from that of the environment ( see also sec .
* boulanger and prault @xcite also derived an estimate of the emission per h atom of the diffuse ism , within 1 kpc of the sun , in the uib range ( 2 to 15 @xmath1 m ) : @xmath100 w.h atom@xmath6 . in this case , eq .
21 gives @xmath122 @xmath14 .
for densities in this range and @xmath80 , @xmath116 is much more sensitive to variations of @xmath65 rather than in @xmath53 .
so , although the fuv content in m 31 may be lower than in the solar environment , as suggested by pagani et al .
@xcite , the weakness of the measured @xmath116 may , in the present model , be due to the relative paucity of hydrogen rather than to that of the fuv flux . when a young , bright star ( e.g. class o or b ) , happens to be in the vicinity or inside a denser ( and hence cooler ) cloud , its fuv radiation digs into the cloud to form an hii region ( signalled by ionic radiation or h recombination radiation ; e.g. br @xmath3 , p @xmath3 ) , followed farther by an hi region ( pdr proper ; essentially h radicals , signalled by centimeter - band radio emission ) , then by a thin layer of radicals recombining into excited molecules ( signalled by several near ir lines ; e.g. 2.42 @xmath1 m ) and , farther still , by the new edge of the molecular cloud ( signalled by co molecular lines ) .
this sequence has often been described ; for a recent analysis , see habart et al .
for our present argument , it is essential to note that , in this type of environment , the uib are observed to come from the intermediate , hi region and the adjacent h@xmath123 region * ( where vibrationally excited h is observed ) : see , for instance , roche et al .
@xcite for the 11.3- and 12.7-@xmath1 m bands , sellgren et al .
@xcite for the 3.3-@xmath1 m band , all observed in the orion bar ; an and sellgren @xcite , studying ngc 7023 , compared the maps of the 3.3-@xmath1 m band with its adjacent , underlying , ir continuum as well as the 1 - 0 s(1 ) line of excited h@xmath0 ; giard et al .
@xcite studied m17 at 3.3 and 10 @xmath1 m , at 6 cm and in the c@xmath124o 2 - 1 band .
this is a very strong indication of a link between h atoms and uib excitation , especially when it is remembered that , in the present model , those molecules that are formed after an h atom impact on a grain , leave the grain surface with a large amount of vibrational excitation ( see sec .
2 ) and that the h@xmath123 near ir emission is therefore generated at the interface between the hi region and the edge of the molecular cloud . *
boulanger et al .
@xcite later analyzed the data collected by the iso satellite from several such locations .
they were thus able to quantitavely determine the brightnesses of the main uibs , at 6.2 , 7.7 , 8.6 , 11.3 and 12.7 @xmath1 m , and this for several sources for which they also determined the illumination @xmath53 .
they found a common trend for all five bands : an initial rise in intensity as a function of illumination , followed by a progressive levelling - off for @xmath125 ( their fig .
the brightness of the 7.7 @xmath1 m band , for instance , starts at @xmath126 w. [email protected]@xmath6 for @xmath80 ( the chamaeleon cirrus cloud in the dffuse ism ) , and terminates at 6.4 w. [email protected]@xmath6 for @xmath127 ( the omega nebula , m 17 ) .
the single uv photon excitation ( also referred to as transient or stochastic heating * in the pah hypothesis ) expects a continuous rise of band brightness with @xmath53 , and the levelling - off is tentatively explained by grain destruction .
the present model predicts an initial linear rise in incompletely dissociated regions followed by a levelling - off as a result of high dissociation , and limited amount , of available h gas . * for a more quantitative comparison of our model with observations , consider the emission @xmath73 given here by eq .
this ir power is emitted in 4@xmath128 sr .
the ir surface brightness of a source of depth @xmath129 along the line of sight is , then , @xmath130 .
the results of astronomical measurements are usually expressed in the form @xmath131 , where @xmath132 .
now , the spectrum of a finite material structure ( or small grain ) , as opposed to that of a homogeneous medium of infinite extent , is expressed in terms of its discrete line intensities , @xmath133 .
there are @xmath134 of these for a total of @xmath135 atoms ( one for each vibrational mode ) ; they can be measured ( absorptivity ) or calculated ( electronic polarizability ) .
if the line distribution is sufficiently dense , an adjacent averaging over @xmath136 makes sense , giving a continuous distribution @xmath137 assuming the excitation energy to be uniformly distributed among all vibrational modes , one can write @xmath138 , and the observed astronomical quantity becomes @xmath139 where @xmath129 is in meters , @xmath73 in w.m@xmath77 and @xmath140 in [email protected]@xmath6 .
finally , a typical value of the pure number @xmath141 can be derived from a bright source , like ngc1482 ( smith et al .
2007 ) , for each of the main bands .
this was done here by interpolating the flux spectrum with 600 points between 5 and 20 @xmath1 m , so that @xmath142 m .
for the 7.65- and 11.3-@xmath1 m bands , for instance , one thus obtains 1.4 and 2.1 respectively . inserting @xmath143 in eq .
( 22 ) , together with expression ( 12 ) for @xmath73 , and @xmath144 ( as in sec .
7.2 ) , we obtain @xmath145 thus , for @xmath146 m ) to comply with the observed value of 10@xmath147 with @xmath80 , @xmath148 should be 3.3@xmath149 @xmath150 with , for instance , @xmath129=1 pc and @xmath151 @xmath14 , which seems reasonable . for this choice , @xmath152 and , in fig .
4 , the representative point falls above the critical line , which would justify the use of approximations 9 and 12 .
it is tempting to use the relation between hydrogen column density and visible extinction along the same line of sight ( if available ) : @xmath153 ( see bohlin et al . 1978 ) . here ,
@xmath154=3@xmath155 , so @xmath156 ; but , in the case of chamaeleon , laureijs et al .
@xcite measured @xmath157=1 .
the discrepancy may be due to a number of causes , the most probable of which is that , for this value of a@xmath158 , the molecular hydrogen fraction in the column density may be quite high due to light extinction by molecules ( see savage et al .
1977 , fig . 6 ; draine and bertoldi @xcite , fig .
8) ; in this case , our expression 3 would underestimate @xmath47 , thus leading to a lower estimate of @xmath65 .
at the high @xmath53 end of boulanger et al.s observations , for m 17 with @xmath127 , @xmath140(7.7@xmath1m)=6@xmath159 [email protected]@xmath6 . equation ( 26 ) then gives @xmath160 cm.@xmath161 .
here , the geometry of the emitting region must be very different than that of a thin cloud , because the very strong illumination by a young , bright , star near by drives a shock in a thick cloud , thus considerably increasing the density in a region of limited extent : a bar " whose thickness along the direction to the star is a small fraction of a parsec ( see , for instance , the extensive study of the orion bar by bregman et al .
@xcite ) . a more likely order of magnitude for @xmath129 ,
therefore , is now 0.1 pc , for which one gets @xmath162@xmath14 , again within reasonable range for this type of object .
more recently , analyzing the data of the _ spitzer _ satellite , smith et al .
@xcite produced a large number of good quality uib spectra from star - forming galaxies .
the peak brightnesses of the main bands cover the range between 10@xmath163 and @xmath164 [email protected]@xmath6 , similar to the galactic data of boulanger et al .
, confirming the remarkable uniformity of the uib * spectral profiles , independant of their intensity .
* a crucial result of boulanger et al.s @xcite observations , among others , is that the relative intensities of the main uibs do not vary much , nor systematically , over a @xmath53 range of @xmath54 .
this insensitiveness was noticed by several authors , observing various environments ( out of galactic plane , other galaxies , etc . ) : see , for instance , murakami et al .
@xcite , chan et al .
@xcite , mattila et al .
@xcite , kahanp et al .
@xcite , smith et al .
it is in agreement with h - impact excitation , which , in nanoparticles , is followed by non - thermal , quasi uniform , redistribution of the same chemical energy ( 5 ev ) among those vibrational modes of the target that are strongly coupled .
by contrast , the insensitiveness to photon flux is at variance with what one would expect from single uv photon excitation as noted by sellgren et al . @xcite . in this model ,
the grain temperature is supposed to decrease with the photon energy so the relative intensities of shorter - wavelength bands would also decrease ; the uibs should even fade out precipitously for illuminating star temperatures below @xmath165 k , as the fuv content of the illumination is then too small .
but this was not borne out by their observations of several reflection nebulae illuminated by stars with temperatures ranging from 5000 to 21000 k ( see also uchida et al .
1998 ) .
mattila et al .
@xcite also remarked that their and others observations implied that the uib carriers are very resistant to different environments ( rne , pne and hii regions ) .
none of the above is incompatible with the small spectral variations in position , intensity or width , which led to the classification of uib spactra in different classes ( see peeters et al .
2002 , tokunaga 1997 ) . in some cases ,
the variations are more radical , such as for the 11 - 13 @xmath1 m massif across ngc7023 ( boulanger et al .
still more striking is the detection or non - detection of the 3.3-@xmath1 m band , independent of illuminating star temperature or ambiant uv content over 4 orders of magnitude of @xmath53 ( smith et al .
this poses a common problem to all models , but appears to be linked with particular chemical processes or environments , rather than with the excitation process ; as such , it falls outside the scope of this work . in most uib spectra ,
an underlying continuum is clearly apparent , rising towards long wavelengths and often very weak at short wavelengths ( see , for instance , cesarsky et al .
2000 , peeters et al .
2004 , zhang and kwok 2010 ) . from their analysis of measurements with the iso satellite in the 6 - 12 @xmath1 m band and at 100 @xmath1 m , kahanp et al .
@xcite concluded that both emissions were strongly correlated at small and large scales , which implies a strong connection ( origin , chemical composition , environment ) between the small grains responsible for the former , and the large ones , responsible for the latter .
mattila et al .
@xcite arrived at the same conclusion from their study of the disk of ngc 891 , which they observed at 8 wavelengths characteristic of different ism components .
this conclusion is in line with the present treatment , as well as with the roughly constant ratio of uib to total ir emission ( sec .
. however , it seems to be at odds with the large variations of the spectral band - to - continuum ratio , or contrast , @xmath166 , from object to object ( see also boersma et al .
if the grain composition does not change with size , this may be explained as follows .
the continuum brightness contribution to the band at @xmath167 can be written @xmath168 where @xmath169 is the grain emissivity , @xmath170 is planck s black - body law ( w.m@xmath112.@xmath1m@xmath6 ) , @xmath129 the object depth along the line of sight ( m ) , @xmath21 the grain number density ( m@xmath77 ) , and @xmath171 the grain optical c - s ( m@xmath172 ) .
the grain temperature is estimated by assuming thermal equilibrium and equating the power absorbed by a grain from the illuminating flux @xmath173 , to the power reemitted by the same grain in the ir . for simplicity , assume the grain absorptivity ( emissivity ) varies as @xmath174 and that all the reemitted power is at the same wavelength , viz .
the peak of the bb ( black body ) emission , @xmath175 , and is given by the stefan - boltzmann law , @xmath176 where @[email protected]@xmath178 . also assume the flux that heats the large grains is @xmath179 , where @xmath53 is the same as above , but @xmath180 is much larger than the dissociating flux , in order to include lower energy photons .
following boulanger and prault @xcite , we take @xmath181w.@xmath18 . also , defining the photon wavelength as 0.1 @xmath1 m , one gets @xmath182 .
then , equating absorbed and reemitted powers , the grain diameter and its emissivity ( at the reference wavelength ) factor away , and it is easily deduced that @xmath183 equation ( 26 ) still requires an estimate of the emissivity , @xmath169 ; it can be obtained from the consideration of the total far ir emission measured by boulanger and prault @xcite in the iras 60- and 100-@xmath1 m bands : @xmath184 w per h atom .
this is to be equated with the stefan - boltzmann law .
now , spitzer @xcite estimated the grain c - s at @xmath185@xmath34 per h atom . assuming the fraction associated with carbonaceous dust to be @xmath186 , then @xmath187@xmath188 .
finally , taking @xmath189k , we obtain @xmath190 at the peak of the black body emission , 150 @xmath1 m and , for an opacity in @xmath174 , @xmath191 using the spitzer result again to express @xmath192 in eq .
26 , the latter becomes @xmath193 finally , the contrast becomes @xmath194 where @xmath73 is taken from eq .
11 and @xmath195 , from eq . 23 . as an illustration , fig .
5 plots @xmath196 as a function of @xmath53 , as well as @xmath166 for @xmath197 m and 11.3 @xmath1 m , both for @xmath198 @xmath14 .
our model predicts that , for most environments in the ism , @xmath53 and @xmath65 are such as to deliver high contrasts , as usually observed .
it also predicts that the contrast decreases as the uib wavelength increases , as observed , except for the diffuse ism ( see onaka et al .
1996 ) : this is because , there , hydrogen is dissociated although the illumination is so low that the larger grains are very cold and the continuum very weak .
finally , the model predicts that the transition from high to low contrast is very swift as @xmath53 increases , due to the steep short wavelength edge of planck s law . as a result ,
the uibs paradoxically disappear " at high illuminations ! in the present model ,
this is not due to their destruction , but to the fact that the underlying continuum quickly increases , even as the uib emission levels off because all available hydrogen is already dissociated ( cf .
observational examples of this behaviour of the band - to - continuum ratio can be found in giard et al .
@xcite ( 3.3-@xmath1 m uib , 10-@xmath1 m continuum ) and cesarsky et al .
@xcite ( 6.2-@xmath1 m uib , 16-@xmath1 m continuum ) , both observing m 17 .
the predictions of our model concur with the conclusions of an and sellgren @xcite , based on their observations of the reflection nebula ngc 7023 with high spatial resolution , in the 3.29-@xmath1 m uib and its neighboring continuum at 2.18 @xmath1 m .
they find the two emissions to coexist in a vast hi region between the hii region enclosing the illuminating star , and the thin h@xmath123 filament at the edge of the molecular cloud .
however , the intensity distributions are distinctly different , with the 3.29-@xmath1 m peaking within the filament , while the 2.18-@xmath1 m peaks at about half this distance from the star .
moreover , they measured a feature - to - continuum ratio peaking at @xmath199 , in the filament , and decreasing with the distance , @xmath94 , from the illuminating star as @xmath200 , i.e. roughly as @xmath201 ( here , @xmath53 is proportional to the star flux heating the grains ) .
remarkably , this trend is very smooth , in stark contrast with the very dissimilar spatial distributions of the two emissions . as noted by the authors ,
both these facts are hardly understandable if both emissions are prompted by absorption of the star radiation . on the other hand ,
the present model predicts just this behaviour of the band contrast
. however , fig .
5 suggests a much faster variation with @xmath53 .
note that the @xmath200 trend is only a regression line over a large set of highly dispersed points , covering only a small range of @xmath53 : a factor 2 , which suggests high density heterogeneity .
so the discrepancy may be due to the lines in fig .
5 being drawn for constant @xmath65 . but it could also be due to a number of assumptions made for the sake of simplicity : -assuming too tight relationships between the densities of large grains ( emitting the continuum ) , uib - emitting grains and h radicals ( sec .
these relationships hold best in the diffuse ism , and may be invalidated in the presence of very intense uv radiation able to drive a shock front into the h cloud . -assuming that the illuminating spectrum does not change with @xmath53 , while , in fact it varies considerably from the diffuse ism to the neighbourhood of a bright star .
ever since pipher @xcite discovered the diffuse ir emission ( dire ) originating in the galactic plane , the problem of its excitation has been open for discussion . based on observations by low et al .
@xcite and rouan et al .
@xcite , respectively , mezger @xcite estimated the total dire at @xmath202w , while ryter and puget @xcite set it at 7.5@xmath203w . given reasonable star formation rates ,
it has been hard to argue that this emission can be accounted for by the luminous o stars alone .
mezger estimated that o stars fuelling radio hii regions contribute only 20 % of the total .
he also concluded that , for uv absorption and ir reemission by the galactic grains in the diffuse medium to account for @xmath204 , these grains should be able to absorb 8 times more uv photons than they can do in fact ( this is the ir excess " ) .
section 6 above suggests that h atom impacts might do the job . in order to explore this possibility in the present model , note that the total number of h atoms in the galaxy was estimated by ryter and puget @xcite and mezger @xcite , again excluding the galactic center , at @xmath205m@xmath206 , which amounts to @xmath207 h atoms . using the values of @xmath204 mentioned above
, we deduce the total ir power reemitted per h atom , @xmath116=4.5 and 14 @xmath99 w / h atom . these are comparable to the value deduced from the iras data by boulanger and prault @xcite : 6.1 @xmath99 .
now , @xmath204 must be compared with the value predicted by our model , i.e. eq .
11 , which gives the total ir power reemitted per unit volume of space , if @xmath71 is replaced by 1 . here , as in sec .
7.1 , we deal with a tenuous environment , so approximation ( 12 ) may be used instead . integrating this quantity over the volume of the galaxy ( excluding star forming regions ) ,
we obtain @xmath208 where @xmath65 is in @xmath14 .
the total volume , v , of the galaxy is the ratio of the total number of h nuclei divided by the average density of the latter .
this is about 1 @xmath14 ( see spitzer 1977 ) , so v@xmath209 @xmath210 . the root mean square density
can then be deduced : @xmath211 where @xmath212 was set at @xmath213 w. the square root of this is 14 @xmath14 , which is typical of clouds containing optically thin h@xmath0 in the dism ( see jura 1975 ) ; the model thus proves to be sound .
the difference between @xmath214 and @xmath215 is a measure of the patchiness of the diffuse ism .
a model was developed for the excitation of the uibs by h atom impacts in the interstellar medium .
it builds upon the fact that , in the presence of far uv radiation and hydrocarbon grains , the hydrogen gas will be partially dissociated and the grain surface will be partially hydrogenated and partially covered with free carbon bonds . under such a statistical equilibrium , h atoms from the gas will recombine with c atoms at the grain surface at some rate . at each recombination , the h atom deposits an energy of about 5 ev in the grain .
half of this is directly converted into vibrational excitation , always distributed in the same way among the most tightly coupled vibration modes of the grain .
absent frequent grain - grain collisions , the only outlet for this energy is ir reemission , part of it in the uibs , provided the chemical structure of the grains is adequate , and the other part in the continuum .
the partition only depends upon the grain size , all grains being assumed to have the same constitution .
only a fraction , @xmath216 , of the grains ( the smallest ones ) will contribute significantly to the uibs .
it is shown quantitatively that h impacts are generally more efficient excitation agents than uv absorption because of the overwhelming abundance of hydrogen relative to uv photons . only very near young bright stars
is this no longer true because photon flux then largely exceeds h atom flux . thus h impacts and fuv absorption are both necessary to understand the variety of observed uib * spectral intensities . *
the model translates into a small number of equations enabling a quantitative comparison of its predictions with available astronomical observations , which have become exquisitely rich and accurate in the last two decades .
section 7 collects some of the most crucial tests of the model : - it predicts uib emission even far from star bursts and in poor - fuv environments . -
since the excitation agent and the deposited energy are always the same , uib reemission is independent of the temperature of the neighbouring illuminating stars . - in pdrs , it becomes possible to understand the observed very near spatial coincidence of hi and h@xmath123 emission with the peak of the uibs .
- the model predicts the measured average power in the uibs ( 2 - 15 @xmath1 m ) emitted per ambient h atom ( in statistical equilibrium in the solar neighbourhood ) , 1.5 @xmath99 w , provided that 1/4 of the cosmic carbon be locked in uib - emitting grains , and the average h nuclei density be about 100 @xmath14 in the average emitting cloud
. - agreement between computed and measured uib intensities is obtained for a given fuv illumination , provided the unknown h column densities are given adequate values , which are found to be quite plausible .
this is true even for m 31 , which is presumed to be poor in fuv flux .
- the assumption of a common carrier for the uibs and the underlying continuum makes it possible to quantitatively account for the observed large variations of the band - to - continuum contrast ratio : they are simply due to the steep variation of wien s law , at a given wavelength , for small variations of illumination .
- the ir excess " paradox , posed long ago by the comparison of the overall galactic ir emission with the uv flux available for grain excitation , is resolved quantitatively by h atom excitation , provided the r.m.s . density of h nuclei throughout the galaxy be of order 200 @xmath217 . * in order to allow quantitative comparisons to be made with observations , a number of assumptions were necessary .
the most restrictive one is the neglect of uv extinction and self - shielding of hydrogen ; the calculations are progressively invalidated as column densities along the sight lines exceed a few times 10@xmath218 @xmath18 .
fortunately , this is not the case for most of the mesured uib spectra . * for lack of relevant experimental results or explicit values from theoretical computations
, i relied heavily on numerical chemical simulation to obtain the values of cross - sections of grain - gas interactions .
as the heating is proportional to the c - h recombination c - s ( eq . 8) , the latter needs measurement or at least confirmation by other means .
it was also assumed that the importance of the dissociating uv flux relative to total uv does not depend on the total uv flux ; this only impacts the degree of dissociation of hydrogen , and most of our calculations are weakly sensitive to it .
none of these assumptions notably affects the general nature and performance of this model . the notion of a fraction
@xmath71 should not be taken to correspond to definite lower and upper size limits ; for the transition between emitting and non - emitting grains is progressive and likely to depend on grain structure and composition .
some of the physical processes involved here are evidenced by molecular dynamic simulations . even for a 100-atom molecule ,
an important broadening mechanism is already at work and directly observable in chemical simulations : this is the interaction between modes and constant redistribution , between them , of the initially deposited energy .
this widens lines into bands and confers wings to each line , so
plateaus " appear below bands " in crowded spectral regions , and even an underlying continuum is visible ( see papoular 2000 ) .
this effect is observed to increase quickly with the size , much more so than the number of modes . as the grain size increases , the number of skeletal vibration modes ( involving the whole structure ) increases more rapidly than the number of modes of functional groups ( involving only a small number of atoms of adequate chemical nature , and with adequate bonds to other atoms ) .
the uibs are mainly due to the latter , while the continuum ( mainly in the farther ir ) is due to the former .
thus , the fraction of energy deposited in the grain that goes into uib emission progressively decreases as the size increases .
this is observed up to the largest structures attainable by the usual simulations ( of order of 10@xmath219 atoms ) .
tailoring the composition and structure of the grains so that they carry the uibs is outside the scope of this paper ( it was treated at length by papoular 2011 ) .
the validity of our conclusions do not depend on these properties , provided the grains bear enough carbon dangling bonds at their periphery , in steady state , so they can capture ambient h atoms .
i thank the reviewer , dr c. boersma , for several useful comments . an j. and sellgren k. 2003 , apj 599 , 312 allamandola l. and norman c. 1978 , a&a 66 , 129 boersma c. , bauschlicher jr .
c. , allamandola l. , ricca a. , peeters e. and tielens a. 2010 , a&a 511 , a32 bohlin r. , savage b. and drake j. 1978 , apj 224 , 132 boulanger f. and prault m. 1988 , apj 330 , 964 boulanger f. , abergel a. , bernard j. , cesarsky d. , puget j .-
, reach w. , ryter c. et al . 1998 , asp conf . ser .
132 , p15 boulanger f. 1999 , les houches lxx , no 11 p253 bregman .
j. , larson k. , rank d. and temi p. 1994
, apj 423 , 326 cazaux s. and tielens a. 2002 , apj lett .
575 , l29 cesarsky d. , lequeux j. , abergel a. et al .
1996 , a&a 315 , l309 cesarsky d. , lequeux j. , ryter c. and grin m. 2000 , a&a 354 , l87 chan k .- w . , rorllig t. , onaka t. et al .
2001 , apj 346 , 273 duley w. and williams d. 2011 , apj lett .
737 , l44 draine b. 1978 , apj suppl .
36 , 595 draine b. and bertoldi f. 1996 , apj 468 , 269 giard m. , bernard j.p . and dennefeld m. 1992 , a&a 264 , 610 guillois o. et al .
1998 , faraday discuss .
109 , 335 graham et al .
1993 , apj 105 , 250 haas m. , klass u. and bianchi s. 2002 , a&a 385 , l23 habart e. , boulanger f. and verstraete l. 2011 , a&a 527 , a122 imamura k. and sofue y. 1997 , a&a 319 , 1 jura m. 1974 , apj 191 , 375 jura m. 1975 , apj 197 , 575 and 581 kahanp j. , mattila k. , lehtinen k. , leinert c. and lemke d. 2003 , a&a 405 , 999 laureijs r. , chlewicki g. , wesselius p. and clark f. 1989 , a&a 220 , 226 lohmar i. , krug j. and biham o. 2009 , a&a 504 , l5 low f. et al .
1977 , apj lett .
214 , l115 low f. et al .
1984 , apj lett .
278 , l19 mattila k. , lehtinen k. and lemke d. 1999 , a&a 342 , 643 mennella v. , brucato j. , colangeli l. and palumbo p. 2002
, apj 569 , 531 mezger p. 1978 , a&a 70 , 565 mezger p. 1982 , a&a 105 , 372 murakami h. , freund m. , ganga k. et al .
1996 , pasj 48 , l41 onaka t. , yamamura i. , tanabe t. , roellig t. and yuen l. 1996 , 48 , l59 pagani l. , lequeux j. , cesarsky d. , donas j. , millard b. , loinard l. and sauvage m. 1999 , a&a 351 , 447 papoular r. 2000 , spectrocimica acta 57a/4 , 947 papoular r. 2005 , mnras 359 , 683 papoular r. 2011 , mnras 415 , 494 ; see also 2008 , arxiv 1008.5136 peeters e. , hony s. , van kerckhoven c. , tielens a. , allamandola l. , hudgins d. and bauschlicher c. 2002 , a&a 390 , 1089 peeters e. , spoon h. and tielens a. 2004 , apj 613 , 986 pipher j. 1973 , iau sympos . 52 radzig a. and smirnov b. 1980 , reference data on atoms , molecules and ions , springer - verelag berlin heidelberg roche p. , aitken d. and smith c. 1989 , mnras 236 , 485 rouan d. , lena p. , puget j .-
de boer k. and wijnbergen , j. 1977 , apj lett .
213 , l39 ryter c. and puget j.l .
1977 , apj 215 , 161 savage b. , bohlin r. , drake j. and budich w. 1977 , apj 216 , 291 sellgren k. , tokunaga a. and nakada y. 1990 , apj 349 , 120 sellgren k. 2001 , spectrochimica acta 57a , 627 smith t. , clayton j. and valencic l. 2004 , astron .
j. 128 , 357 smith j. , draine b. , dale d. et al .
2007 , apj 656 , 770 solomon p. , carangelo r. , best p. , markham j. and hamblen d. 1986 , 21st sympos . on combustion , 437 spitzer l. 1977 , physical processe in the interstellar medium , j. wiley & sons , n. y. stecher t. and williams
d. 1979 , a&a 67 , 115 tokunaga a. 1997 , asp conf .
vol 124 , 149 turro n. et al .
2009 , principles of molecular photochemistry uchida k. , sellgren k. and werner m. 1998 , apj lett .
493 , l109 zhang y. , kwok s. and hrivnak b. 2010 , apj 725 , 990 | a model was developed for the excitation of the uibs by h atom impacts in the interstellar medium .
it builds upon the fact that , in the presence of far uv radiation and hydrocarbon grains , the hydrogen gas will be partially dissociated and the grain surface will be partially hydrogenated and partially covered with free carbon bonds . under such a statistical equilibrium , h atoms from the gas will recombine with c atoms at the grain surface at some rate . at each recombination , the h atom deposits an energy of about 5 ev in the grain .
half of this is directly converted into vibrational excitation , always distributed in the same way among the most tightly coupled vibration modes of the grain .
absent frequent grain - grain collisions , the only outlet for this energy is ir reemission , part of it in the uibs , provided the chemical structure of the grains is adequate , and the other part in the continuum .
the partition only depends upon the grain size , all grains being assumed to have the same constitution . only a fraction , about 0.25 , of the grains ( among the smallest ones ) will contribute significantly to the uibs .
it is shown quantitatively that h impacts are generally more efficient excitation agents than uv absorption because of the overwhelming abundance of hydrogen relative to uv photons .
only very close to young bright stars is this no longer true because photon flux then largely exceeds h atom flux .
thus h impacts and fuv absorption are both necessary to understand the variety of observed uib spectra .
the model translates into a small number of equations enabling a quantitative comparison of its predictions with available astronomical observations , which have become exquisitely rich and accurate in the last two decades .
[ firstpage ] |
despite the fact that self - organization of a system in an out - of - equilibrium state plays a crucial role in dynamical structural formation in nature , physical principles behind such phenomena have remained elusive .
active matter @xcite and driven granular matter @xcite are two important classes of out - of - equilibrium systems .
they share an intrinsic out - of - equilibrium nature , and the only basic difference is that the energy is injected locally for the active systems whereas globally for the granular systems @xcite .
this global nature of energy input makes granular matter physically simpler than active matter .
thus , granular matter is not only important for its own sake , but also regarded as a model for understanding the physics of active matter .
granular matter is an important class of materials , distinct from thermal systems since the thermal energy is negligible for its description .
granular matter is ubiquitous in nature and its dynamical self - organization always takes place in a strongly non - equilibrium situation as in active matter , since energy input is essential for its occurrence @xcite .
its statistical yet athermal nature makes the physical description extremely difficult . from an experimental point of view
, the control of self - organization of granular matter is also a difficult task .
however , a notable exception is a dynamic steady state , maintained by the balance between energy input and dissipation , which allows us to perform well - controlled experiments .
the most idealized system may be a quasi - two - dimensional ( 2d ) driven granular particle monolayer , where spherical particles are confined between two parallel plates whose gap is narrow enough to avoid particle overlap along the vertical direction and energy is injected by vertically vibrating plates .
this system allows us to access all phase - space information at the particle level .
so the phase behavior of such a monolayer particle system has played a crucial role in our understanding of the fundamental nature of self - organization in a system far from equilibrium .
this vibrated monolayer particle system has also attracted considerable attention for its connections with fundamental problems in the field of condensed matter and statistical physics @xcite . the liquid - solid transition in a 2d disk system , the thermodynamic counterpart of a vibrated monolayer , has been a hot topic since the discovery of the liquid - solid transition for hard disks by alder and wainwright @xcite .
two - dimensional particle systems can not crystallize at finite temperature due to significant fluctuation effects associated with the low dimensionality , yet the above work shows that they may still form solids .
there is a long - standing debate @xcite on the nature of this transition for a system of the simplest interparticle interaction , hard disks .
one scenario is that ordering takes place via two steps , liquid - to - hexatic and hexatic - to - solid transitions , now widely known as the kosterlitz - thouless - halperin - nelson - young ( kthny ) scenario @xcite .
here each transition is continuous .
the other is that ordering takes place in one step via a first - order liquid - solid transition @xcite .
there have been hot debates on which is the relevant scenario .
very recently , it was shown @xcite that the transition actually takes place by a scenario different from both of them : it occurs with two steps as the kthny scenario suggests , but the first transition is not continuous but weakly discontinuous .
however , the first - order nature of the liquid - hexatic transition is very weak , and the transition roughly obeys the kthny scenario .
this basic behavior is common to other systems including particles interacting with soft repulsive potentials @xcite and those with attractive potentials such as the lennard - jones potential @xcite , although it has recently be shown that the nature of the transitions depends on the softness of the potential in a delicate manner @xcite .
monolayer granular matter has provided a model experimental system to study this fundamental problem .
some time ago , careful experiments were made on the athermal counterpart of the above system .
it was shown by shattuck and his coworkers that a driven monolayer particle system continuously transforms from a liquid to an intermediate hexatic , and then to a solid phase , with an increase in the particle area fraction @xmath0 under a constant @xmath1 @xcite ( see sec .
ii for the precise definitions of @xmath1 and @xmath0 ) .
a similar meting transition behavior was also observed by olasfen and urbach when increasing the dimensionless acceleration @xmath1 at a fixed particle area fraction @xmath0 for a granular quasi - monolayer @xcite .
however , it was shown that the thickness of the cell @xmath2 , which is 1.6 times of the particle diameter @xmath3 , plays a crucial role in the transition : height fluctuations of particles may be a source of disorder .
the increase of their amplitude with an increase in the vibration amplitude , or @xmath1 , increases the number density of defects , eventually leading to the melting of the solid phase .
thus , the mechanism may be essentially different from the former example , which does not involve any significant hight fluctuations due to a strong 2d confinement .
the former liquid - solid transition behavior as a function of @xmath0 @xcite obeys the kthny scenario @xcite , although the liquid - to - hexatic transition may be weakly first - order @xcite .
this study suggests that a quasi-2d driven granular system behaves very similarly to its thermal counterpart .
a similar conclusion was also derived for glass - transition - like phenomena of driven binary mixtures @xcite and polydisperse systems @xcite .
the energy injected by mechanical vibration is converted to the kinetic energy of a system and the effective ( granular ) temperature @xmath4 is defined by this kinetic energy .
if this is high enough to overcome the gravity and the energy loss originating from the friction and inelastic collisions with the container @xcite , we may approximately regard the system as a thermal equilibrium system as long as interparticle collisions are almost elastic .
however , the exact mechanism responsible for this apparent thermal behavior of an athermal system has remained elusive .
we note that these experiments were performed by using rather elastic balls such as steel balls .
then , the natural question to ask next is how the nature of the liquid - solid transition is affected by the inelasticity of collisions , or internal dissipation .
there were pioneering works on liquid - solid transitions of quasi-2d granular systems @xcite .
these studies showed interesting monolayer liquid - bilayer solid coexistence in a nonequilibrium steady state , in which the two phases have different granular temperatures .
these are intriguing examples of phase ordering , more generally self - organization , in a dynamic steady state of a non - equilibrium open system , maintained by the balance between energy input and dissipation . here
we study the effect of energy dissipation on liquid - solid transition of a quasi-2d driven granular matter by comparing the behaviors of steel and rubber ball systems .
we note that steel and rubber balls have differences in the restitution coefficient , the friction coefficient , and the elastic properties . among these
, the difference in elastic properties may be less important compared to the others because forces acting upon interparticle collisions are too weak to cause non - linear deformation of the balls and it is known that the 2d melting behavior is not affected by the softness of the interaction potential @xcite .
a situation we consider is a _ single _ layer of monodisperse spheres , which is vibrated between two horizontal plates @xcite , unlike the above - mentioned previous works where bilayer formation is allowed @xcite .
the control parameters in our experiments are the energy input characterized by a dimensionless acceleration @xmath1 and the area fraction of particles @xmath0 . here
we demonstrate that the dissipation due to the inelasticity of collisions and the friction can , if they are strong enough , completely break the similarity between the athermal and the corresponding thermal system and fundamentally change the nature of the liquid - solid transition _ in a monolayer _ from the kthny - like continuous transitional behavior @xcite to a strongly discontinuous transition .
we discuss a physical mechanism responsible for this unconventional self - organization under energy dissipation .
our study reveals a novel mechanism leading to the coexistence of two phases with different granular temperatures , which is essentially different from the mechanism previously found @xcite .
we infer that a similar mechanism may be relevant to self - organization of active matter including living systems .
dissipative interactions , such as inelastic , hydrodynamic , frictional interactions , may play a crucial role in the state selection .
our experimental apparatus is analogous to those used in @xcite and the same as that used in @xcite .
a monolayer of non - cohesive particles was vibrated sinusoidally in the vertical direction by an electromagnetic shaker ( labworks et-139 ) with frequency @xmath5 .
we changed the dimensionless acceleration @xmath6 , where @xmath7 is the amplitude of vibration and @xmath8 is the gravitational acceleration , by controlling @xmath7 .
the cell has a circular shape and its inner diameter is @xmath9 .
the circular annulus made of duralumin is used as the hard side wall .
when we make an experiment with a soft side wall ( see sec .
ii.d ) , the inner side wall of the annulus is covered by a silicone rubber sheet of 0.2 mm thickness . in our experiments
, we use two types of spherical particles : steel and fluorine - rubber balls .
the particle diameter is @xmath10 for both stainless steel and fluorine - rubber ball .
the fluorine - rubber ball has four important characteristics : ( i ) a non - cohesive character , ( ii ) a large stiffness ( @xmath11 gpa ) , which allows us to ignore nonlinear deformation upon collision , ( iii ) a low dynamic friction coefficient ( @xmath12 ) , and ( iv ) a low restitution coefficient . the friction coefficient of the fluorine rubber may be comparable to that of the steel . however , the coefficient of rolling friction may be smaller for the steel ball than for the rubber ball since the former has a smoother surface than the latter .
the restitution coefficient @xmath13 is difficult to estimate accurately and it is known to depend on the particle velocity .
here we measured the height ratio before and after a collision with a steel wall by dropping a ball vertically .
the restitution coefficient @xmath13 against a steel wall was estimated as @xmath14 and @xmath15 with the variance of 0.1 for the steel and rubber ball respectively .
we also estimated the @xmath13 for collision between two rubber balls as @xmath16 .
the value of @xmath13 for steel balls coincides well with the literature value @xcite . by confining these particles between two plates separated by an annulus with a thickness @xmath2 of 4.0 mm ( i.e. , @xmath17 ) , we allowed only quasi-2d particle motion , i.e , no bilayer formation .
we note that in previous similar works on the liquid - solid transition of driven granular matter @xcite @xmath18 , which allows particles to form bilayers ( see appendix a-1 ) .
this difference in @xmath19 leads to a crucial difference in the nature of a liquid - solid transition between our system and theirs , as will be discussed later .
the top plate is a transparent glass plate so that we can observe the particles through it .
the bottom duralumin plate is covered with a sandpaper by which the vertically oscillating particles are scattered @xcite .
the use of a wall with surface roughness is expected to randomize the horizontal motion and realize brownian - like motion of particles .
it was shown @xcite that a large enough @xmath1 leads to the velocity distribution function of a nearly gaussian shape for a driven monolayer system .
this is because the randomization of the energy injection due to particle collisions and the wall roughness leads to realization of a self - generated effective white bath in a long timescale . for steel balls
, we obtained the particle coordinates by tracking the bright spots at the top of particles reflecting illuminating light . for rubber balls , on the other hand , we obtain the particle coordinates simply by binarisation of particle images .
we used a high speed camera ( vcc - h500 , digimo co. ltd . ) .
the typical frame rate used was 100 frame / s and the image resolution was 640@xmath20480 .
occasionally , we used the rate of 500 frame / s with a resolution of 640@xmath2090 .
a pixel size corresponds to 0.25 mm .
the position of each particle was tracked by a particle tracking software , which fits a gaussian function to an image of each particle .
the detection error is less than 0.05 mm for a steel ball and 0.1 mm for a rubber ball .
we measured a structure after attaining a steady state ( typically after 10 min from the initiation of vibrational driving ) . in our system ,
the transition between a solid - like and liquid - like state occurs as a function of the area fraction @xmath21 , where @xmath3 is the diameter of particle , @xmath9 is the inner diameter of the annulus , @xmath22 is the total number of particles in the system .
we note that this definition of @xmath0 is often used in describing the phase transition of ( quasi-)2d hard particle systems ( see , e.g. , @xcite ) .
we note that the random close packing and the closest packing in 2d occur at about 0.83 and 0.906 respectively .
because of the rather small size of our system , our measurements may suffer from finite - size effects .
the effects may not be so serious for a discontinuous liquid - solid transition in a rubber ball system , since there is no diverging lengthscale .
on the other hand , the results of a steel ball system may suffer from finite - size effects .
however , the maximum correlation length we could attain was 9 times the particle diameter @xmath3 , which is still much shorter than the cell diameter ( @xmath23 ) .
so the finite size effects may not be so severe for our results , although there might be slight shifts in the transition area fractions .
this conclusion is supported by the rather good agreement of the measured @xmath0-dependences of the correlation length of hexatic order @xmath24 and translational order @xmath25 with the theoretical predictions for an infinite system ( see below ) .
however , the correlation lengths , @xmath24 and @xmath25 , may be bound by the cell size very near the hexatic ordering point @xmath26 and the solidification point @xmath27 , respectively , and the weak discontinuous nature of a liquid - hexatic transition may be smeared out by the finite - size effects @xcite . in principle
, we can use a cell with a larger diameter , but what is most important in our measurements is the homogeneity of the vertical vibration amplitude .
so we used only the rather small cell .
we are planning to make a large cell with high mechanical rigidity , but we leave this for future investigation . the 2d radial distribution function @xmath28 was calculated as @xmath29 which is the ratio of the ensemble average of the number density of particles existing in the region @xmath30 to the average number density @xmath31 . here @xmath22 is the number of particles in the simulation box , whose side length is @xmath32 , and @xmath33 is the increment of @xmath34 .
the spatial correlation of translational order is characterized by the translational correlation length @xmath25 for @xmath35 as follows : @xmath36 where @xmath37 expresses an operation to extract the envelop function of the positive part of a function @xmath38 .
the hexatic order parameter is measured by @xmath39 for each particle @xmath40 , where the sum runs over the @xmath41 nearest - neighbors of particle @xmath40 , and @xmath42 is the angle between the bond @xmath43 and a fixed arbitrary axis . here
the nearest - neighbor particles are detected by a criterion @xmath44 .
we confirmed that this criterion provides the same nearest - neighbor identification as a method based on voronoi tessellation .
then , the spatial correlation of @xmath45 is calculated as @xcite @xmath46 the spatial correlation of the bond - orientational order can then be characterized by @xmath47 . here
the division by @xmath28 is to remove the effect of translational ordering . to characterize the fluctuations of @xmath48
, we estimate the spatial correlation length of @xmath48 , @xmath24 , and the susceptibility , @xmath49 .
we obtain the spatial correlation length @xmath24 for @xmath50 from the following relation : @xmath51
now we show experimental data which provide information on the nature of the phase transitions . for driven steel balls , it was previously shown that the system transforms from the liquid to solid via the intermediate hexatic phase @xcite .
for comparing the behavior of steel and rubber ball systems on the same ground we performed experiments for the two systems under the same experimental conditions . before showing results , we briefly review how the spatial correlations of the positional and bond orientational order parameter are predicted to grow with an increase in the area fraction @xmath0 in the framework of the kthny theory @xcite .
the predicted behaviors are summarized in fig .
[ fig : kthny ] . the hexatic ordering point @xmath52 and the crystallization point @xmath53
are characterized by the power - law decays of the bond orientational and positional order parameter , respectively .
the former should obey @xmath54 at @xmath52 , whereas the latter @xmath55 at @xmath27 .
we first describe the results of a steel ball system .
we show the @xmath0-dependence of @xmath28 and @xmath56 for a steel ball system in fig .
[ fig : one ] .
the spatial decays of both quantities change from an exponential to a power - law decay , we identify the ordering points by whether the decay is slower than the predicted power law decay or not . because of the limitation of our system size ,
the firm confirmation of the exponent of asymptotic power - law decay of these correlation functions are difficult .
but the results are at least consistent with the prediction of the kthny scenario ( see also below ) . as shown in fig .
[ fig : kthny ] , @xmath57 should decay slower than @xmath54 for @xmath58 , whereas @xmath59 should decay slower than @xmath55 for @xmath60 . in figs .
[ fig : one](a ) and ( b ) , these lines with a slope of @xmath61 and @xmath62 are drawn as guides to estimate @xmath52 and @xmath53 respectively . in this way
we determine @xmath63 and @xmath64 .
the determinations of @xmath52 and @xmath53 are further supported by the divergence of the correlation length and the susceptibility of the hexatic order parameter for the former and by the divergence of the translational order correlation length for the latter .
the slight differences of the transition area fractions from those of a hard disk system @xcite may be due to the quasi-2d nature of a system and finite size effects . as examples of our analysis , here
we show details of our fittings of @xmath56 for the steel ball system , respectively , in fig .
[ fig : supple_xi6 ] . the fittings to @xmath65 are basically the same . from these analyses ,
we obtain the @xmath0-dependence of the spatial correlation length of the hexatic order @xmath48 , @xmath24 , and that of the translational order , @xmath25 , although there are considerable errors due to the small system size .
the growth of these lengths with an increase in @xmath0 are used to confirm the prediction of the kthny theory in a quantitative manner ( see fig . [
fig : xi_chi](a ) and ( b ) ) .
we can also see indications that the decays of the envelops of @xmath56 and @xmath65 change from exponential to power law , respectively , around @xmath66 and @xmath67 .
the phase boundaries between liquid , hexatic , and solid phases are determined by standard methods used to study a thermal system , such as the nature of the decay of the spatial correlation of bond orientational and translational order ( see fig .
[ fig : kthny ] ) .
we measured the correlation length of the hexatic order parameter @xmath48 , @xmath24 , as a function of the area fraction @xmath0 .
firstly , the decay of the correlation in the hexatic order parameter changes from an exponential to a power - law decay at the liquid - hexatic transition point , @xmath26 ( @xmath68 ) ; secondly , the density correlation changes from an exponential to a power - law decay at the hexatic - solid transition point @xmath27 ( @xmath69 ) ( see above ) . as shown in fig .
[ fig : xi_chi](a ) , we observe the steep divergence of @xmath24 towards @xmath26 , which is consistent with the prediction of the kthny theory , @xmath70 $ ] .
we also fit the relation of @xmath71 to the data of @xmath25 , although the accuracy of the data is not so high .
we also observe a sharp peak in the susceptibility of @xmath48 , @xmath49 at @xmath26 ( fig .
[ fig : xi_chi](c ) ) . in this way
, we confirm the presence of a liquid , hexatic , and solid state and determine the phase - transition compositions @xmath26 and @xmath27 .
however , the estimation of @xmath27 may not be so accurate since the range of the analysis of @xmath65 is limited by finite size effects ( see sec .
furthermore , the short - range order developing for @xmath72 do not allow us to access an asymptotic power law decay of @xmath65 in a wide distance range . to avoid finite size effects
, we need to perform experiments in a large cell
. however , we can at least see the very slow algebraic decay of @xmath65 above @xmath27 .
we note that , for steel balls , we do not observe any indication of liquid - solid coexistence , also suggesting the ( nearly ) continuous nature of the two transitions .
the basic phase ordering behavior of a steel ball system is thus fully consistent with the kthny scenario , as reported previously by shuttack and his coworkers @xcite . ,
( filled circle ) towards the hexatic ordering point @xmath26 and that of the translational order , @xmath25 , ( open circle ) towards the solidification point @xmath27 for a steel ball system .
the solid and dashed curves are the prediction of the kthny theory ( see fig .
[ fig : kthny ] ) .
( b ) the @xmath0-dependence of @xmath24 ( filled circle ) and @xmath25 ( open circle ) for a homogeneous liquid state of a rubber ball system .
( c ) the susceptibility of the hexatic order parameter @xmath73 for steel ( blue filled circle ) and rubber ball systems ( red filled circle ) .
the steep increase near @xmath26 is observed for a steel ball system , reflecting the nearly continuous nature of the transition , whereas there is no such behavior for a rubber ball system , reflecting the strong discontinuous nature of the transition .
( d ) the area fraction of the solid phase to the total area , @xmath74 , as a function of @xmath0 .
it starts to increase from @xmath75 until @xmath76 in proportional to @xmath77 , indicating the validity of the lever rule .
we also confirmed the similar relation between the number of particles of the solid phase @xmath78 and the total number of particles @xmath22 .
this means that each coexistence phase retains the same number densities irrespective of @xmath0 in the coexistence region .
the error bars in ( a)-(c ) represent the standard deviations of the fittings .
for all the data , @xmath79 ] next , we focus on the liquid - solid transition in a rubber ball system . unlike in the above steel ball system
, we observe in this system the two - phase coexistence of the liquid and solid phases separated by a rather sharp interface at a certain range of @xmath0 ( see fig .
[ fig : coexistence ] ) .
the two - phase coexistence can also be clearly seen from the bimodal shape of the probability distribution function of the hexatic order parameter @xmath80 , as shown in fig .
[ fig : probability ] .
below @xmath75 , the spatial correlations of hexatic and translational order are both short - range and decayed nearly exponentially , as shown in fig . [
fig : rubber_g(r ) ] . as can be seen in fig .
[ fig : xi_chi](b ) , neither the translational nor orientational correlation length , @xmath25 and @xmath24 , exhibit any growth with an increase in @xmath0 , unlike the case of a steel ball system .
there is also no increase in the susceptibility around the phase transition points , as shown in fig .
[ fig : xi_chi](c ) .
since the analysis of the decay of the spatial correlation functions is not useful in the coexistence region , we distinguish a liquid state , a solid state , and a coexistence state on the basis of measurements of the local orientational order , the local density ( or , area fraction ) , and the local mobility , which has a link to the granular temperature , ( see fig . [
fig : coexistence ] ) .
we can see these three different quantities including both static and dynamical ones can identify the two - phase coexistence in a consistent manner , although the analysis of an instantaneous structure leads to some inaccuracy due to short - time fluctuations of the interface ( see below on a possible origin ) .
we can also clearly see that the fraction of the solid phase monotonically increases with an increase in @xmath0 .
the area fraction of the solid phase determined from the hexatic order parameter is shown in fig .
[ fig : xi_chi](d ) , which indicates the validity of the lever rule ( see below ) .
a discontinuous first - order - like phase transition often accompanies hysteresis and metastability .
we made all the observation after attaining a steady state ( typically after 10 min from the initiation of vibrational driving ) and the behavior was very reproducible .
we did not observe any indication of hysteresis mainly because measurements are always done after a steady state is reached and there is no way to change the area fraction continuously .
so we identify @xmath75 and @xmath76 as the lower and upper boundary of the solid and liquid phase , respectively .
= 0.70 , 0.72 , and 0.74 , which are between @xmath75(=0.695 ) and @xmath76(=0.775 ) , as @xmath79 .
the two - phase coexistence can be seen by binarisation by using the following three quantities : ( a ) the hexatic order parameter @xmath48 of each particle ( here the threshold is chosen as @xmath81 , and the pattern is insensitive to the choice for @xmath82 ( see fig .
[ fig : probability ] ) ) .
yellow ( solid ) and black ( liquid ) particles correspond to particles having @xmath83 and @xmath84 , respectively .
( b ) the local density @xmath85 , which is calculated from the voronoi area of each particle ( the threshold @xmath86 mm@xmath87 ) .
blue - green ( solid ) and pink ( liquid ) particles correspond to particles having @xmath88 and @xmath89 , respectively .
( c ) the displacement over 10 s of each particle ( the threshold @xmath90=3.4 mm ) .
blue ( solid ) and green ( liquid ) particles correspond to particles having @xmath91 and @xmath92 , respectively .
we can see that the solid regions identified by these three quantities are well correlated with each other .
we can also see the monotonic increase of the solid fraction with an increase in @xmath0 .
, width=321 ] -dependence of the probability distribution function of the hexatic order parameter @xmath48 , @xmath80 , at @xmath79 .
blue curves are for a steel ball system , whereas red ones for a rubber ball system . for a steel ball system
, @xmath93 always has a unimodal shape , whereas for a rubber ball system it has a clear bimodal shape for @xmath0 between @xmath75 and @xmath76 .
the long tails of @xmath80 toward low @xmath48 in homogeneous solid phases come from defects . , width=321 ] here we show examples of of our fittings of @xmath28 and @xmath56 for the rubber ball system at @xmath94 in fig . [
fig : rubber ] .
we find no systematic @xmath0-dependences for @xmath28 and @xmath56 ( see fig .
[ fig : xi_chi](b ) ) .
we can see in fig .
[ fig : rubber ] that even at @xmath94 , which is very near @xmath95 , the decays of density correlation and hexatic order correlation are both exponential and much faster than the power law decays of exponent @xmath62 and @xmath61 , respectively . above @xmath76
, we see a homogeneous solid phase . for @xmath96 ,
we observe the coexistence of the liquid and the solid phases .
we confirm that the former has the upper bound @xmath0 of the liquid phase , @xmath75 , and the latter has the lower bound @xmath0 of the solid phase , @xmath76 .
we determine @xmath95 and @xmath97 .
the area fraction of each phase obeys the lever rule in the coexistence region , as shown in fig .
[ fig : xi_chi](d ) . the strong discontinuous nature of the transition
is also confirmed by that fact that there is neither the divergence of @xmath24 ( fig .
[ fig : xi_chi](b ) ) nor the sharp increase in @xmath73 ( fig .
[ fig : xi_chi](c ) ) at the phase boundary .
we also show the distribution function of @xmath48 , @xmath80 , in fig .
[ fig : coexistence](e ) and ( f ) .
we can see a clear bimodal distribution for the coexistence region of @xmath0 , which is another clear indication of the liquid - solid coexistence .
such a signature of the strongly first - order transition is absent for a system of steel balls .
we find that the coexistence can be seen not only by the local packing symmetry characterised by the hexatic order parameter ( fig .
[ fig : coexistence](a ) ) , but also by the area fraction @xmath0 ( fig .
[ fig : coexistence](b ) ) and the in - plane displacement amplitude ( fig . [ fig : coexistence](c ) ) .
the last point indicates that the effective ( granular ) temperature @xmath4 defined by the kinetic energy is spatially inhomogeneous for the liquid - solid coexisting state .
thus , the solidity is linked not only to the static quantities such as the area fraction and the bond orientational and translational order , but also to the effective temperature .
this is a very unique feature of this non - equilibrium steady state , which is maintained by continuous vibrational energy input and dissipation due to inelastic interparticle collisions .
this phenomenon has some similarity to the inhomogeneization of a granular gas due to inelastic collisions @xcite .
coexistence of phases with different granular temperatures was also reported for a quasi-2d system where bilayer formation is allowed @xcite , however , the underlying mechanism may be quite different , as will be discussed later .
the link between the solidity and the effective temperature indicates that the interfacial profile may be related to the spatial gradient of not only the area fraction and the hexatic order but also the temperature ( i.e. , the kinetic energy ) .
this possible dependence of the interfacial profile on the mobility , or the effective temperature , is unique to non - equilibrium open systems and absent in thermal equilibrium systems ( see , e.g. , @xcite ) . on the basis of these results ,
we draw the phase ( more strictly , state ) diagrams for steel and rubber balls , which are shown in fig . [
fig : pd ] . here
we include the dependence of the phase behavior on @xmath1 .
in general , we need @xmath1 larger than a critical value @xmath98 to maintain a dynamical steady state . below @xmath98 ,
the energy injection becomes inhomogeneous .
the gray region labeled inhomogeneous excitation " in fig . [ fig : pd](b ) is in such an inhomogeneous state .
this is because a well - defined dynamical steady state can be maintained only when the energy injected by vertical vibration overcomes the effects of the gravity and the energy loss due to inelasticity of collisions of balls and the walls @xcite ( see also appendix a-2 and appendix b ) .
we find the critical value of @xmath1 to maintain a dynamical steady state is higher for the rubber ball system ( @xmath99 ) than for the steel ball system ( @xmath100 ) , reflecting the stronger inelastic nature of collisions of the former with the confining plates .
we note that the transition area fractions , @xmath26 , @xmath27 , @xmath75 , and @xmath76 , are all independent of @xmath1 within the accuracy of our measurements .
the insets of fig .
[ fig : pd](a ) and ( b ) show the local orientational order in the solid phases formed in a steel and rubber ball system at @xmath101 , respectively .
we can clearly notice that the amount of defects is much larger for a rubber ball system than for a steel one .
this can also be seen in the shape of @xmath80 , which has a larger tail toward low @xmath48 for a rubber ball system than for a steel one .
this may be just a consequence of the difference in the lower stability limit @xmath0 of the solid phase between a steel and a rubber ball system , but there might be other fundamental reasons . in the range studied .
( b ) rubber ball systems , which show the distinct discontinuous transition between liquid and solid states . in this case
, the dynamical steady state is realized only above @xmath102 .
this higher threshold value of @xmath1 is presumably due to the inelastic nature of collisions of rubber balls with the confining plates . the images in ( a ) and ( b ) are configurations in the solid phase at @xmath101 for steel and rubber ball systems , respectively .
green , blue , and red particles have local hexagonal , pentagonal , and heptagonal structures , respectively . the inner part surrounded by the red dashed circle
has less order ( i.e. , more defects ) for the rubber ball system than for the steel one .
, width=302 ] we also note that the solid phase is always formed in the middle part of the container far from the side wall ( see fig .
[ fig : coexistence ] and fig .
[ fig : boundary](a ) ) . this may be explained by a larger restitution coefficient of particle - side wall collision than particle - particle one .
the hard side wall prefers the liquid phase with high @xmath4 .
we confirm that a softer side wall ( covered by a silicone rubber film ) is statistically more wettable to the solid phase ( fig .
[ fig : boundary](b ) ) , although wetting is rather modest likely because the curved wall inevitably induces elastic distortion of the solid phase .
thus , this wetting phenomenon may be regarded as _ dissipation - induced wetting_. .
the liquid phase preferentially wets the wall .
( b ) the same as ( a ) for a softer wall whose surface is covered by a silicone rubber film . in this case , the solid phase partially wets the wall ( see the regions surrounded by the red lines ) . ]
finally , we briefly mention a related interesting phenomenon we observe in a mixture of steel and rubber balls .
there are some studies on inelasticity - induced demixing @xcite ; however , there is no example associated with a liquid - solid transition . here
we report inelasticity - induced demixing , which is linked to the formation of an ordered solid phase .
when we mix a small amount of steel balls with rubber balls , we observe steel balls are completely expelled from the solid region and included in the liquid region ( see fig . [
fig : mixture](a ) ) .
we note that the solid phase is made of only rubber balls : _ dissipation - induced demixing_. here we stress that in such a mixture the energy input rate to steel balls is higher than that to rubber balls . on noting that steel balls have a larger kinetic energy due to its larger restitution coefficient
, the above argument naturally explains the preferential inclusion of steel balls into the liquid phase .
if the fraction of steel balls is too large , the coexistence conditions can not be satisfied simultaneously and thus the formation of the macroscopic ordered solid phase is largely prohibited even at the same area fraction @xmath0 . as a result ,
a rather random mixed liquid state is realized ( see fig . [
fig : mixture](b ) ) . ) driven at @xmath79 . here
we can see a coexistence of the solid phase purely made of rubber balls and the liquid phase which is a mixture of steel and rubber balls .
( b ) snapshot of a 1:1 mixture of steel and rubber balls ( the total area fraction of @xmath103 ) driven at @xmath79 .
the color of the outer shell of each particle represents the type of the particle ( blue : steel ball ; red : rubber ball ) , whereas that of the inner core the degree of the hexatic order ( black : low @xmath48 ; yellow : high @xmath48 ) . ]
here we consider the relationship of our results with those of pioneering works on phase orderings in quasi-2d driven granular systems @xcite , in which bilayer formation is allowed .
we note that there is a crucial difference in the dependence of the phase behavior on the strength of inelasticity : for systems , where a bilayer can be formed , ordering ( towards bilayer solid ) is suppressed by inelasticity @xcite .
we emphasize that this is completely the opposite to our case : inelasticity helps ordering ( towards monolayer solid ) .
see appendix a-1 for a more detailed discussion including an intuitive explanation on the cause of the difference .
we also note that the monolayer - bilayer formation can take place as a function of @xmath1 @xcite , whereas our transition can not be induced by changing @xmath1 as we can see from the phase diagrams in fig .
[ fig : pd ] . near @xmath98
, we see a crossover from inhomogeneous excitation to homogeneous one , but this transition as a function of @xmath1 has an origin essentially different from the transition as a function of @xmath0 ( see appendix a-2 and b for the details ) .
the above - mentioned fundamental differences between our work and the previous studies strongly indicate that the underlying physics is essentially different between the two cases even on a qualitative level . in appendix a-3
, we also mention another type of inhomogeneization in driven inelastic particles , gas - liquid coexistence @xcite .
finally , we mention the 2d melting of a solid phase with an increase in @xmath1 @xcite .
as described in the introduction , this melting is induced by the increase in the defect density , which is caused by the increase in hight fluctuations of particles .
since our 2d constraint @xmath104 is much stronger than the one used in this work ( @xmath105 ) , we did not see any indication of @xmath1-induced melting .
it is interesting to study in the future how and at which @xmath19 the behavior changes .
first we qualitatively consider what is the origin of the difference in the phase - transition behavior between the steel and rubber ball systems .
steel and rubber balls differ not only in the inelasticity and friction upon interparticle collisions but also in the softness of interparticle interactions .
it is rather well established that the kthny - like behavior , or a transition from liquid to solid via the intermediate hexatic phase , is very robust for 2d ` thermal ' systems , irrespective of the types of interparticle interactions : essentially the same behavior was observed not only for hard disks but also for particles interacting with soft repulsive potentials and attractive ( e.g. , lennard - jones ) potentials ( see , e.g. , refs .
@xcite ) , although it has recently been shown that the liquid - hexatic transition depends delicately on the softness of the interaction @xcite .
this robustness of the kthny - like behavior irrespective of the nature of interactions is a natural consequence of the fact that hexatic order is formed primarily by geometrical packing effects : when hard - sphere - like particles are packed , this local configuration is entropically favored .
this suggests that the nature of the liquid - solid transition between the steel and rubber ball systems is not controlled by the softness of particles , but by the dissipation due to the inelasticity of particles and the friction .
now we consider how inelasticity and friction can change the nature of the liquid - solid transition .
steel balls are not perfectly elastic but the system still behaves like a thermal system . on the other hand ,
the behavior of the rubber ball system is distinctly different from the behavior of its thermal counterpart .
so the question can be rephrased as what physical mechanism controls the transition from apparently thermal to athermal behavior when we change the degree of inelasticity and friction from steel to rubber balls . for a vertically vibrated monolayer system , the particle-(top and bottom ) wall collision frequency , @xmath106 ,
is higher than the particle - particle collision frequency , @xmath107 , as long as @xmath108 .
we estimated @xmath106 as roughly 100 hz by illuminating light from a low angle from the horizontal plane and @xmath107 as roughly 20 - 50 hz , depending upon on the local @xmath0 , by using normal illumination from the top , with a fast camera .
for an inelastic particle system to apparently behave like an elastic thermal system , the energy dissipated by each particle - particle collision should be fully recovered by energy input through particle - wall collisions before the next collision with a particle takes place .
as long as this condition is satisfied , even a dissipative system apparently behaves like a thermal system , as our steel ball system does .
the athermal nature is accompanied by inhomogeneization of a system and the resulting coexistence of two phases with different effective granular temperatures
. such two - phase coexistence characteristic to athermal systems was already discovered for a quasi-2d driven granular system , in which bilayer formation is allowed unlike our strictly monolayer system @xcite . in such a case ,
the different granular temperatures can be explained by the difference between monolayer excitation and bilayer excitation ( e.g. , differences in effective mass and type of excitation ) .
however , this mechanism can not explain the two - phase coexistence we found in a driven monolayer system .
below we consider the origin of inhomogeneization and the resulting coexistence of two phases with different granular temperatures on a qualitative level . here
we consider a gedanken experiment , where the inelasticity is gradually increased in a continuous manner . with an increase in the inelasticity of particles , the perfect recovery before
the next interparticle collision becomes more and more difficult . with an increase in the particle density , or @xmath0 , the inter - particle collision frequency @xmath107 , which controls the rate of energy dissipation , increases but with a rather constant particle - wall collision frequency @xmath106 , which makes this recovery process less efficient .
the energy dissipation is due to both the inelastic and the frictional nature of interparticle collisions .
we believe that this crossover from the perfect to imperfect recovery of the kinetic energy during @xmath109 with an increase in the particle density destabilizes a non - equilibrium steady state with spatially homogeneous density and granular temperature , leading to the transition from an apparently thermal to strongly athermal behavior .
now we qualitatively consider how the imperfect recovery makes a system inhomogeneous : in such a situation , the degree of recovery of the kinetic energy of a particle should decrease with an increase in local @xmath0 since @xmath107 increases with an increase in @xmath0 .
thus , higher density regions formed by spontaneous fluctuations should transiently have lower kinetic energy than lower density regions , which leads to lower pressure in the former .
this mechanism is essentially the same as that of clustering instability in dissipative gases @xcite .
so the higher density regions are further compressed , resulting in the enhancement of density fluctuations .
however , an increase in density eventually causes the increase in the inter - particle collision frequency and thus the increase in pressure . when the horizontal pressure is spatially homogeneized , this development of density fluctuations stops , ending in two - phase coexistence . in this way the system finally reaches a steady state , where the energy input rate and the dissipation rate is balanced while keeping two phases with different @xmath0 and granular temperature @xmath4
thus , a high enough density region and a low enough density region should coexist in a steady state for a certain range of the average @xmath0
. the average horizontal pressure in the two phases should be the same since the mechanical force balance is to be satisfied across the domain interface .
because of its mechanical nature , this equal pressure condition may be hold even in an athermal condition ( see below , however , on the effect of interparticle friction )
. however , the fluctuations of pressure may be much larger than in a thermal system , since the particle velocity is determined not only by interparticle collisions but also by particle - wall collisions .
the another constraint comes from the condition for a steady state : the balance between the energy input rate and the dissipation rate should be balanced in each phase
. the low density liquid phase is characterized by low @xmath107 , which means low energy dissipation and results in high @xmath4 , whereas the high density solid phase by high @xmath107 , which results in low @xmath4 .
this situation is realized under a homogeneous energy input , or spatially homogeneous @xmath106 . on the basis of this physical picture ,
below we consider a principle behind the two - phase coexistence on a phenomenological level . here
we note that it is interesting to study the dependence of the phase - transition behavior as a function of inelasticity . however , a precise control of particle inelasticity is not easy in experiments , and thus numerical simulations may be more promising . before discussing a principle behind the two - phase coexisting more theoretically ,
here we consider which theoretical framework is suitable for the description of what we discussed above .
one candidate is a hydrodynamic theory for a confined granular system , which was recently developed by brito et al .
it was shown that under a situation where the thermostat does not inject momentum but only energy , the equations for the conserved density field and momentum density are the continuum equation and the navier - stokes one , as in the case of usual fluids , and the equation for the non - conserved ( granular ) temperature field is a balance equation for the energy .
a collisional model for a 2d system was also developed on the same basis ( see sec .
iv of ref .
@xcite ) ; however , in this model the stationary temperature is density independent and the pressure increases monotonically with density , and accordingly there is no two - phase coexistence , contrary to our experimental observation . in this model , the granular temperature in a homogeneous stationary state was assumed to be determined by a balance between energy dissipation and injection @xcite , but the energy dissipation due to an effective friction proportional to a particle velocity was not considered .
although it is interesting to take this effect into account in this theory and to study whether it can induce two - phase coexistence , here we take a more phenomenological approach . to take this velocity - dependent dissipative force into account in a natural manner
, we consider a thermal condition to maintain a steady state in a quasi-2d system , on the basis of a langevin - like equation of motion ( see , e.g. , @xcite ) .
now we consider a principle behind two - phase coexistence in a non - equilibrium steady state . for a thermodynamic system , in order for two phases of the same substance in contact to be in equilibrium , there must be mechanical , thermal , and chemical equilibrium , i.e. , ( i ) equal pressure , ( ii ) equal temperature , and ( iii ) equal chemical potential . for our driven granular system , however , condition ( ii ) is not be applicable because of the out - of - equilibrium nature .
on the basis of a physical picture we described in sec .
iv.b , here we focus our attention on this condition , using a model of granular brownian motion @xcite .
the change in particle velocity due to a binary instantaneous collision between particle @xmath110 and @xmath40 with the same mass @xmath111 is given by @xmath112 \hat{\mathbf{n}}$ ] , @xmath113 \hat{\mathbf{n}}$ ] , where @xmath114 and @xmath115 are the velocity after and before the collision respectively , @xmath116 is the unit vector jointing the centres of particles and @xmath13 is the restitution coefficient ( @xmath117 ) which is equal to 1 in the elastic case . in order to maintain a steady state ,
we need an external energy source that is coupled to every particle in the form of a thermal bath . in our case
, this external energy is supplied by collisions with vibrating walls confining a granular monolayer .
the motion of a particle @xmath110 is then described by the following stochastic equation @xcite : @xmath118 here @xmath119 is the force taking into account the collisions with other particles , @xmath120 is a drag coefficient characterizing the velocity decay towards a steady state , whose timescale we express by @xmath121 , and @xmath122 is random white force noise exerted by particle - wall collisions , with @xmath123 and @xmath124 ( @xmath125 being the effective bath temperature ) . here
we note that the viscous term in eq .
( [ eq : bd ] ) takes into account the friction among particles and energy transfers between various degrees of freedom ( e.g , the friction between particles and the walls ) @xcite .
here we mention the random nature of force noise . in our system , the excitation itself has a well - defined frequency , but the randomization of the energy injection by interparticle collision and the wall roughness leads to a self generated effective white bath in a long timescale .
the langevin - type approach can be rationalized by the fact that brownian dynamics simulations can fully reproduce the behavior of a driven monolayer granular system of weakly polydisperse steel balls even without any adjustable parameters @xcite .
its validity to a rubber ball system is not obvious , but our visual inspection of the motion of steel and rubber balls at least indicates the same type of random motion .
although the exact nature should be investigated , e.g. , by measuring the velocity distribution function in the future , this visual inspection and the consideration in sec .
vi.c tell us that in a steady state the physics may be the same between the two systems at least on a phenomenological level .
so we assume that the randomization also takes place efficiently for a rubber ball system .
a stationary state is maintained since the effect of the external energy source balances the energy lost by interparticle collisions and particle - wall collisions .
the key parameters of the system is the characteristic velocity decay time , @xmath121 , and the packing fraction @xmath0 . multiplying the above stochastic equation eq .
( [ eq : bd ] ) by @xmath114 and averaging yields @xmath126 here we note that the granular temperature @xmath4 is defined as @xmath127 and thus the left - hand side of the above equation can be written as @xmath128 .
the first term in the right - hand side of eq .
( [ eq : bd2 ] ) can be written as @xmath129 , where @xmath130 ^ 2 $ ] , represents the average energy dissipation rate due to interparticle collisions .
@xmath131 is the collision average , whose expression is known for granular gas @xcite but not for high - density liquid or solid because of the difficulty associated with not only excluded volume effects but also recollisions and memory effects .
we express this term @xmath129 as @xmath132 .
although its exact form is not clear , it should be a decreasing function of @xmath13 and an increasing function of @xmath0 , and @xmath4 .
the second term is rewritten from the definition of @xmath4 as @xmath133 . finally , the third term is the total energy input rate and written as @xmath134 .
here we briefly consider the role of interparticle friction .
although the friction coefficient itself may be similar between the steel and rubber balls , the rubber ball may have a larger coefficient of rolling friction with the walls as well as another ball upon collision than the steel ball does . on a phenomenological level
, this effect can be included in the coefficient @xmath135 ; then , @xmath135 , or @xmath121 , should be an increasing function of @xmath0 .
the friction also affects the term @xmath136 ( see , e.g. , ref .
we also note that the presence of interparticle friction also induces forces tangential to the domain interface .
this may perturb the interface position and induce large fluctuations of the interface .
such a signature can indeed be seen in fig .
[ fig : coexistence ] , although it might be largely due to the large pressure fluctuations unique to a driven granular system ( see sec .
iv.b ) . for simplicity
, however , we do not consider this dynamical effect when we discuss the phase coexistence . although the difference in the friction property between steel and rubber balls may play a role in the observed difference in the phase transition behavior ,
the coefficient of rolling friction is generally small for smooth spherical particles .
thus , we argue that the difference in the restitution coefficient between the steel and rubber ball should be the major source of the difference in the energy dissipation between the two systems . by incorporating all the above factors ,
( [ eq : bd2 ] ) can be expressed as follow : @xmath137 in a steady state , we should have the relation @xmath138 .
this relation is just a consequence of energy conservation , simply implying that the energy input to a 2d granular system is partially dissipated by inelastic interparticle collisions and by viscous damping due to the particle - wall interactions . for an elastic system
where @xmath139 and thus @xmath140 , this relation together with @xmath138 reduces to the relation @xmath141 . strictly speaking
, we should also consider the energy flux arising from the spatial inhomogeneity of @xmath0 and @xmath4 ( see below ) , but we tentatively ignore it here since it is not relevant to the description od a steady state . on the basis of the above physical picture , we argue that the strong discontinuous nature of the transition of an inelastic system found here is a consequence of the inhomogenization of a system under the above - mentioned constraint @xmath142 . here
we assume that @xmath143 is a function of the two state variables , @xmath0 and @xmath4 .
first the condition of equal pressure @xmath144 immediately tells us that @xmath145 since @xmath146 , which is seen in fig .
[ fig : coexistence](d ) .
we note that this condition may be robust since it is of mechanical origin , as described before .
then the two conditions @xmath147 , together with eq .
( [ eq : e ] ) , tell us that the denser solid phase dissipates more energy .
another condition is the chemical equilibrium condition , which is also a subtle issue .
chemical - balance conditions of two coexisting phases with different temperatures should be obtained as a steady state solution of the relevant kinetic equations also considering bond orientational and translational ordering .
it may be a promising way to introduce a non - equilibrium free energy @xcite , @xmath148 , and consider the balance of @xmath149 between the two phases , which are expected to be the necessary conditions to maintain a steady state .
our discussion is purely phenomenological , and furthermore we also need to consider @xmath48 .
thus , a more rigorous approach is highly desirable in the future .
the lever rule @xmath150 is then obtained from the condition ( 1 ) @xmath151 , i.e. , @xmath152 , where @xmath153 , @xmath154 , and @xmath155 , are the particle area fraction , the number of particles , and the total area of phase @xmath110 , and ( 2 ) @xmath156 . the similar lever rule was also observed for the monolayer - bilayer transition ( see , e.g. , @xcite ) .
the crucial difference from a thermal system arises from the fact that in our system the granular temperature ( i.e. , the kinetic energy ) , @xmath4 , is different between the two phases .
the above condition @xmath138 is required for maintaining the dynamical steady state dissipating energy and it may be this condition that plays the most crucial role in the phase selection in a non - equilibrium state .
the side wall - particle collisions dissipate energy with a rate different from bulk , which affects the wettability of a phase to a solid side wall , as shown in fig .
[ fig : boundary ] . from eq .
( [ eq : e ] ) , we can infer that a wall harder than particles tends to wet a liquid phase of high @xmath4 rather than a solid phase of low @xmath4 whereas a wall softer than particles tends to wet a solid phase rather than a liquid phase .
this is simply because particles near a hard / soft side wall tend to have higher / lower @xmath4 .
this spatial inhomogeneity of @xmath4 near a side wall should be coupled to the local area fraction @xmath0 to satisfy the following steady state condition for the balance of the heat flux @xmath157 @xcite : @xmath158 where @xmath159 and @xmath160 are transport coefficients .
this condition qualitatively explains why the solid phase with a higher @xmath0 tends to wet to the softer wall , which locally lowers @xmath4 . demixing shown in fig .
[ fig : mixture ] can be explained by the fact that the energy loss upon collision with the top and bottom plates is smaller for steel balls than for rubber balls .
the resulting higher kinetic energy of steel balls is a reason why steel balls tends to be located in a liquid phase of hight @xmath4 .
this is again consistent with the physical principle discussed above .
a higher kinetic energy of a steel ball should lead to a larger specific area per particle for it .
such a tendency can be seen in fig .
[ fig : mixture](a ) and ( b ) .
this can also be explained by the condition of the balance of the heat flux @xmath157 ( see eq .
( [ eq : q ] ) ) , which tells us the negative correlation between the local effective temperature @xmath4 and the local area fraction @xmath0 .
to summarize , we find that energy dissipation plays a crucial role in self - organization of driven monolayer granular matter , which allows the coexistence of states with different effective temperatures , contrary to the phase coexistence in a thermal system .
the two - phase coexistence obeys the lever rule as in a thermodynamic first - order transition ; however , the underlying selection rule is fundamentally different in the sense that the energy dissipation rate , which is an intrinsically non - equilibrium quantity , is the key factor of the phase selection . since our discussion is phenomenological , however , it is desirable to theoretically describe the coexistence conditions in a non - equilibrium steady state in a more rigorous manner .
there are also many fundamental open questions such as what determines the interface profile and what is the nature of fluctuations of the interface .
we also show that it is possible to separate particle species by solidification using the inelasticity contrast , which is similar to _ purification _ of materials including impurities by crystallization in thermodynamic systems .
our findings may shed light on a general principle governing the state selection of granular matter far from equilibrium , which should also be important for our understanding of industrial processing of granular materials by vibration and flow .
finally we expect that a similar principle may hold for the state selection in active matter , which is another important class of out - of - equilibrium systems . for active systems ,
there have recently been many studies on a liquid - solid transition @xcite , glass transition @xcite , demixing of self - propelled particles @xcite and rotors @xcite .
these studies have elucidated unique characters of the phenomena distinct from their thermodynamic counterparts .
it has been clarified that the coupling between local energy input ( or motility ) and density @xcite plays a crucial role in the state selection ( see , e.g. , @xcite ) .
inclusion of inelastic interactions , nonlocal viscous dissipation , and local friction may also alter the nature of the state selection in such a system significantly @xcite .
in active matter , and particularly in living systems , interactions between active objects are often dissipative or non - conservative .
thus , it is quite interesting and important to study effects of dissipative interactions such as inelasticity and local ( or nonlocal ) friction on the state selection of active matter in a systematic manner .
we hope that our study can aid the understanding of fundamental roles of dissipative interactions in self - organization of out - of - equilibrium systems .
we thank john russo and taiki yanagishima for critical reading of our manuscript .
this study was partly supported by grants - in - aid for scientific research ( s ) and specially promoted research from the japan society for the promotion of science ( jsps ) .
here we compare the results of our work with those of previous works done in a quasi-2d situation , for which bilayer formation is allowed @xcite .
all the previous works , which reported liquid - solid coexistence , were made for a cell height @xmath2 of about 1.7 - 2.0 times of the particle diameter @xmath3 .
accordingly , these systems can form bilayers , and the liquid - solid transition in these works always accompanies monolayer ( liquid)-bilayer ( solid ) transition . thus
, the situation is essentially different from ours , where the cell thickness is thin enough to avoid bilayer formation .
interestingly , this extra degree of freedom , bilayer formation , makes the physics essentially different from our case in which granular particles always form a monolayer .
the crucial difference can be seen most clearly in the dependence of the phase behavior on the strength of inelasticity .
urbach and his coworkers reported that inelasticity significantly expands the low - density liquid region for a system in which bilayer formation is allowed @xcite .
furthermore , computer simulations showed that the ordered phase is not present at any vibration amplitude when the inelasticity is large .
similar behaviors were also reported by clerc et al .
these results clearly indicate that _ ordering is suppressed by inelasticity_. we stress that this is completely the opposite to our case : in our case inelasticity changes the nature of the transition from continuous to discontinuous and the upper bound density of a disordered liquid state is lower for the inelastic system than for the elastic one , i.e. , _ inelasticity helps the ordering_. another important difference comes from the fact that bilayer formation leads to a change in the number density of the bottom layer because of the conservation of the total number of particles .
this extra degree of freedom also causes a difference in the physics between the two types of systems .
lobkovsky et al .
@xcite showed by numerical simulations that the monolayer - bilayer transition becomes rather continuous for random forcing .
this is an interesting observation in the sense that the way of driving affects the nature of the transition .
however , they also showed that inelasticity suppresses the onset of the ordered phase with random forcing , as is observed in the vibrating system .
we stress that this tendency is again the opposite to our case .
furthermore , we note that the monolayer - bilayer formation can take place as a function of @xmath1 @xcite , whereas our transition _ can not _ be induced by changing @xmath1 , as we can see from the phase diagrams in fig .
[ fig : pd ] . a change from a discontinuous to a continuous nature of the @xmath1-induced transition
was observed with an increase in the normalized cell thickness @xmath19 from 1.83 to 1.94 @xcite .
here the key control parameter is the cell thickness , which affects the symmetry of the bilayer solid phase ( square or hexatic symmetry ) @xcite , and not the degree of inelasticity .
this fact and the above - mentioned crucial difference in the @xmath1-dependence of the transition also tell us that the physics is essentially different between the two cases .
the phase ordering behavior in granular matter was proposed to be generally expressed by the ginzburg - landau - type free energy ( or potential ) and relevant equations of motion @xcite . on the basis of this picture , a theoretical explanation for bilayer formation was proposed by clerc et al . @xcite .
the thermodynamic force on the density field @xmath161 mainly comes from the effective ( non - equilibrium ) pressure gradient calculated from this potential .
the authors also included the friction term for the @xmath161 field and random force noises .
the kinetic equation governing the phenomena was shown to be non - diffusive and include not only @xmath162 ( i.e , @xmath163 ) but also @xmath164 ( i.e. , @xmath165 ) @xcite .
the sum of these forces leads to acceleration of the field @xmath161 .
the reaction - diffusion - type equation of motion with these forces captures the travelling wave features .
this is markedly different from our monolayer case , where we have never observed such travelling waves .
we confirmed by comparing results of experiments and those of brownian dynamics simulations that our driven monolayer granular system obeys langevin ( or brownian ) dynamics with viscous damping and random noise @xcite .
this indicates that there is no acceleration term , which is proportional to @xmath165 , in our system .
thus , the above theory can not be applied to our monolayer case .
this presence or absence of the acceleration term crucially affects the dynamics and make our monolayer system distinct from a bilayer - forming system .
nevertheless , it does not influence a steady state , since there @xmath166 for any quantity @xmath167 , in a steady state , thus , the two - phase coexistence is determined by the ginzburg - landau - type potential @xcite .
this coexistence is then controlled solely by the parameter @xmath168 , which is a coefficient of the quadratic term ( @xmath169 ) in the potential .
this parameter @xmath168 was assumed to be proportional to the inverse of the compressibility coefficient , as in the case of a usual thermodynamic gas - liquid coexistence .
this ( negative ) compressibility tells us how easily bilayers can be formed . in this model ,
the coexistence is determined by the equal pressure and the equal non - equilibrium chemical potential under the mass conservation .
the parameter @xmath168 was shown to be given by the derivative of the momentum flux with respect to the density and its negative value , which leads to the coexistence , reflects the fact that the granular temperature is lower for a higher density @xcite .
thus , this non - equilibrium chemical potential allows the granular temperature to be different between the two coexisting phases .
here we propose an intuitive explanation for the role of inelasticity in the liquid - solid transition accompanying bilayer formation , i.e. , a relation between the above @xmath168 and inelasticity .
for a quasi-2d system which can form a bilayer , even for elastic particles there is a liquid - solid coexistence ( see , e.g. , fig . 3 in @xcite ) .
this is because the bilayer formation should be easier for more elastic particles since a smaller loss of the kinetic energy associated with interparticle collisions allows particles in the bottom layer to more efficiently jump to the top layer . with an increase in the inelasticity ,
the discontinuous nature becomes weaker and eventually the transition becomes continuous for particles having a small enough restitution coefficient .
this termination of the discontinuous first - order - like transition was called a critical point ( @xmath170 ) by clerc et al .
criticality was also observed experimentally , using @xmath1 as a control parameter , when @xmath1 approaches a critical value @xcite .
the theory may be valid for a quasi-2d driven granular system where bilayer formation is allowed .
however , we emphasize again that the above - mentioned dependence on the inelasticity is completely opposite to what we find in our system : our central finding is that an increase in inelasticity changes the nature of the 2d liquid - solid transition from ( thermal - like ) two - step continuous transitions to a one - step discontinuous transition . for bilayer formation
the kinetic energy has to overcome both gravity and energy loss due to inelastic collisions .
this tells us that more elastic particles can form a bilayer more easily , as described above . to our knowledge
, our work is the first to show a discontinuous first - order - like liquid - solid transition of a monolayer granular system in a high @xmath1 regime , whose nature is primarily controlled by energy dissipation , as discussed in sec .
next , we mention another ` apparently ' similar behavior , but which also has a very different physical origin .
olafsen and urbach @xcite reported clustering or ordering upon decreasing the vertical vibration amplitude for a quasi-2d system , but which has no upper plate and thus is not confined by two parallel walls unlike our case : at large @xmath1 , particle correlations exhibit only short - range order as in the case of equilibrium 2d hard - sphere gases , but lowering @xmath1 cools the system , resulting in a dramatic increase in correlations leading to either clustering or an ordered state .
further cooling forms a collapse : a condensate of motionless balls coexisting with a less dense gas . measured velocity distributions are non - gaussian , showing nearly exponential tails . in our systems
, we observed similar phenomena below a critical value of @xmath98 , which was shown in fig .
[ fig : pd](b ) for a rubber system as the `` inhomogeneous excitation '' state ( the gray region ) . for a steel ball system
this phenomenon was observed for a much lower value of @xmath1 , as described in the main text .
we note that these phenomena have an essentially different physical origin , which is for example discussed in detail on the basis of molecular dynamics ( md ) simulation @xcite and a navier - stokes granular hydrodynamics @xcite .
nie et al . showed by md simulation @xcite that at high @xmath1 the particle motion is isotropic , and the velocity distributions are gaussian .
the deviations from a gaussian distribution at low @xmath1 is related to the degree of anisotropy in the motion . below @xmath98 ,
the vertical velocity distribution becomes bimodal : the cluster particles move with the plate , while the gas particles are non - interacting , as they collide primarily with the plate .
they proposed that dissipative contact forces are responsible for this phenomenon .
it was also mentioned by khain and aranson @xcite that the phenomenon can be viewed as a consequence of a negative compressibility of granular gas .
this explanation is somewhat similar to that for bilayer formation discussed above @xcite , but the range of @xmath1 is very different between the two : the clustering is observed for low @xmath1 , whereas the ordering accompanying bilayer formation for much higher @xmath1 . in relation to this , it was stated @xcite that the behavior does not significantly depend on the inelasticity of collisions between the particles ; one does not need inelastic particle collisions to reproduce experimental observations .
they proposed that the mechanism of phase separation occurring at low @xmath1 is related to the non - trivial interplay between the energy injection and the vertical temperature of the particles . in any case , thus , an inhomogeneous granular temperature is primarily a consequence of inhomogeneous energy injection .
it was also shown that the behavior does not significantly depend on the inelasticity of collisions between the particles .
this clearly indicates that this phenomenon has a physical origin essentially different from ours .
finally , we mention gas - liquid coexistence observed in driven inelastic particles @xcite . in this case
, the difference in the type of particle motion between the two phases was found to play a key role in the coexistence : in the dense liquid phase , the injected energy is quickly dissipated within the bulk by frequent interparticle collisions due to a high density . in the dilute gas phase , on the other hand , the motion of particles is synchronized with the driving , which reduces the relative velocity between particles and thus the rate of interparticle collisions .
thus , two phases with a large difference in the granular temperature coexist .
however , this phenomenon is also essentially different from ours , reflecting the difference in the cell thickness ( many layers vs. monolayer ) and the nature of the transition ( gas - liquid vs. liquid - solid transition ) ; for example , there is no such synchronized motion in our system .
recently brito et al .
@xcite showed that for a granular monolayer vibrated between the walls , uniform energy input to a system is generally a good assumption . in a quasi-2d geometry ,
in which only a monolayer can exist , the system is known to remain homogeneous in the horizontal directions for a wide range of parameters .
this is due to the presence of a distributed energy source .
they also pointed out that in the absence of friction , this energy source is galilean invariant and conserves momentum locally . in such a quasi-2d system ,
the vertical energy scale of grains is fixed by vibration parameters .
we stress that the energy injection occurs only through direct collisions of a particle with the walls , and there is no other channel .
this means that the energy injection rate is controlled by @xmath106 .
so , as long as there are no direct geometrical restrictions to the vertical motion of the particles , it is reasonable to assume that the energy input is rather homogeneous spatially .
we note that , according to our observation , there is no overlap of particle images projected onto the horizontal plane ( see , e.g. , images in fig . [
fig : coexistence ] ) . for steel balls ,
the input energy is homogeneous up to @xmath171 , which is much higher than the upper bound area fraction of the liquid phase , @xmath53 ( = 0.695 ) , of the rubber ball system .
importantly , as shown in fig .
[ fig : pd ] , all the phase transition area fractions , @xmath26 , @xmath27 , @xmath75 , and @xmath76 , are independent of @xmath1 .
this suggests that the particle - wall collision frequency @xmath106 may be almost the same between the two phases , or rather homogeneous spatially , for this range of @xmath1 .
we note that if the particle - wall collision frequency @xmath106 ( i.e. , a vibrational parameter ) strongly depends on @xmath0 , the phase boundary compositions should also depend on @xmath1 .
thus , the energy input rate may be assumed to be homogeneous .
however , since the discussion above is qualitative , the spatial distribution of the energy input in two - phase coexistence needs to be checked carefully by numerical simulations in the future . for a quasi-2d system , in which a bilayer can be formed ,
the situation is very different .
for example , it was clearly shown @xcite by mapping the local average rate of energy input that a square phase consisting of bilayer corresponds to a region of dramatically reduced energy input .
however , this is natural since the bilayer formation itself inevitably accompanies a change in vibrational parameters ; for example , the effective mass of the vibrated object is roughly doubled by bilayer formation .
we note that such effects are absent for a monolayer system , as mentioned above .
65ifxundefined [ 1 ] ifx#1 ifnum [ 1 ] # 1firstoftwo secondoftwo ifx [ 1 ] # 1firstoftwo secondoftwo `` `` # 1''''@noop [ 0]secondoftwosanitize@url [ 0 ]
+ 12$12 & 12#1212_12%12@startlink[1]@endlink[0]@bib@innerbibempty @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop _ _ ( , ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop ( ) @noop ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) link:\doibase 10.1103/physreve.77.061301 [ * * , ( ) ] @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) link:\doibase 10.1103/physrevlett.114.188301 [ * * , ( ) ] @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) | self - organization of active matter as well as driven granular matter in non - equilibrium dynamical states has attracted considerable attention not only from the fundamental and application viewpoints but also as a model to understand the occurrence of such phenomena in nature .
these systems share common features originating from their intrinsically out - of - equilibrium nature .
it remains elusive how energy dissipation affects the state selection in such non - equilibrium states . as a simple model system , we consider a non - equilibrium stationary state maintained by continuous energy input , relevant to industrial processing of granular materials by vibration and/or flow .
more specifically , we experimentally study roles of dissipation in self - organization of a driven granular particle monolayer .
we find that the introduction of strong inelasticity entirely changes the nature of the liquid - solid transition from two - step ( nearly ) continuous transitions ( liquid - hexatic - solid ) to a strongly discontinuous first - order - like one ( liquid - solid ) , where the two phases with different effective temperatures can coexist , unlike thermal systems , under a balance between energy input and dissipation .
our finding indicates a pivotal role of energy dissipation and suggests a novel principle in the self - organization of systems far from equilibrium .
a similar principle may apply to active matter , which is another important class of out - of - equilibrium systems . on noting that interaction forces in active matter , and particularly in living systems , are often non - conservative and dissipative , our finding may also shed new light on the state selection in these systems . |
the observation of gravitational waves ( gws ) by advanced laser interferometer gravitational wave observatory ( aligo ) @xcite has initiated a new era of exploring the cosmology , the nature of gravity as well as the fundamental particle physics by the gws detectors @xcite . especially , due the the limitation of the colliders energy , gws detectors can be used as new techniques to probe the symmetry breaking patterns or phase transition history for a large class of new physics models with hidden gauge group extensions of the standard model ( sm ) , which are motivated by the experimental results in our understanding of particle cosmology and fundamental physics ( such as the dark matter mystery or the puzzling observed baryon asymmetry of the universe ) , and no evidence for new physics discovered at current collider experiments .
the increasingly attractive new physics models with hidden non - abelian gauge group extensions of the sm have many new particles without leaving any observable imprints at current particle colliders .
however , the gws experiments may provide a possible approach to test their existence .
for example , to explain the baryon asymmetry of the universe via electroweak baryogenesis by the hidden sectors , a strong first - order phase transition is usually needed to realize the departure from thermal equilibrium @xcite . and
during the first - order phase transition , the gws will be produced through three mechanisms : collisions of expanding bubbles , sounds waves , and magnetohydrodynamic turbulence of bubbles in the hot plasma @xcite .
phase transitions in particle physics and cosmology are usually associated with the symmetry breaking , i.e. where the universe transits from a symmetric phase to a symmetry broken phase when the temperature drops below the corresponding critical temperature .
for the first time , we have a chance to explore the hidden gauge symmetry breaking through phase transition gws signals after the discovery of the gws by aligo , which is is particularly exciting . in this paper , we study the possibility to to explore the hidden gauge symmetry breaking patterns by the phase transition gws signals .
in particular , we will focalize our analysis to gws detection of the non - abelian gauge group extended models , where the symmetry breaking at each energy scale may associate with a first - order phase transition , as shown in fig.[sb ] . the hidden group @xmath0 , which can spontaneously breaks into the sm gauge group @xmath1 . during the hidden gauge symmetry breaking ,
a strong first - order phase transition can take place , which will produce detectable phase transition gws .
for example , the hidden gauge group @xmath0 can be the non - abelian group @xmath2 , and the model is called 3 - 3 - 1 model @xcite .
we show that many versions of the 3 - 3 - 1 model can produce at least one time strong first - order phase transitions at the energy of several tev in some parameter spaces , which can produce detectable gws spectrum at evolved laser interferometer space antenna ( elisa ) @xcite , big bang observer ( bbo ) @xcite
. in general , there can exist many times spontaneous symmetry breaking , which may also accompany several times first - order phase transition as shown in fig.[sb ] .
if the energy scale of the symmetry breaking and first - order phase transition is at @xmath3 gev triggered in the hidden sector , the phase transition gws spectrum is within the sensitivity of future aligo .
other attractive examples of gws detection of the non - abelian gauge group extended models are also discussed , such as the first - order dark qcd phase transition gws in the relaxion mechanism @xcite , which may be tested by pulsa time array ( pta ) at the square kilometre array ( ska ) @xcite or the five - hundred - meter aperture spherical telescope ( fast ) @xcite in china . the other one is the nnaturalness mechanism @xcite , which may also produce observable phase transition gws signals at the gws detector . this paper is organized as follows : in section ii , we schematically discuss the gws detection of the hidden gauge group symmetry breaking and show how to calculate the phase transitino gws during the first - order phase transition . in section iii , we will study the phase transition gws spectra in some concrete hidden gauge group extended models . in section iv , we show our final discussions and conclusions .
in a large class of new physics models with extended hidden non - abelian gauge group , one or several times cosmological phase transitions can occur during each step s hidden gauge symmetry breaking .
in general , we can have the gauge symmetry breaking patterns with the extended hidden gauge group at each scales with the evolution of our history as shown in fig .
for example , the hidden symmetry breaking pattern may be the following : @xmath4 with the evolution of our universe , symmetry breaking will happen at corresponding energy scale , where the strong first - order phase transition may take place .
detailed models are given in section iii .
along with the symmetry breaking from a ` false ' vacuum to a ` true ' vacuum , strong first - order phase transition occurs if there exists a sufficient potential barrier between them .
this process can produce an observable stochastic gws signals , which can be detected in some gws detectors , such as aligo , elisa @xcite , bbo @xcite , ska , fast and so on .
their sensitivity range for some critical temperatures depends on the energy scale of the first - order phase transition for different dark hidden group extended models , as we show in section iii . to discuss the gws spectrum from the first - order phase transition
, it is needed to begin with the one - loop finite temperature effective potential : @xmath5 where @xmath6 represents the order parameter of the phase transition ( a real scalar field ) , @xmath7 is the one - loop coleman - weinberg potential at @xmath8 , and @xmath9 is the thermal contribution including the daisy resummation @xcite . during each step of symmetry breaking in the hidden gauge group extended models ,
strong first - order phase transition may occur . during the first - order phase transition
, bubbles are nucleated via quantum tunneling or thermally fluctuating the potential barrier .
the bubble nucleation rate per unit volume @xmath10 is given by @xmath11 with @xmath12 @xcite , where @xmath13 is the euclidean action @xcite defined as @xmath14 .
\label{eq_bounces}\end{aligned}\ ] ] here , @xmath15 , and @xmath16 @xcite where @xmath17 .
\label{eq_bounces3}\end{aligned}\ ] ] from the above equations , in order to obtain the bubble nucleation rate , the profile of the scalar field @xmath6 during the bubble nucleation is needed to be calculated by solving the following bounce equation : @xmath18 with the boundary conditions @xmath19 which can be solved numerically using the overshoot / undershoot method .
the first - order phase transition terminates when nucleation probability of one bubble per horizon volume is of @xmath20 , i.e. , @xmath21 .
that is to say , it should satisfy @xmath22 it is known that there exist three major sources for producing gws during the first - order phase transition , which are collisions of the vacuum bubbles @xcite , sound waves @xcite and turbulence @xcite in the plasma after collisions , respectively .
the most well - known source is the bubbles collisions , and the first - order phase transition gws spectrum depends on four parameters .
the first one is the ratio @xmath23 of the vacuum energy density released in the phase transition to that of the thermal bath , defined as @xmath24 where @xmath25 specifies that the quantity is evaluated at the time @xmath26 determined by eq.([tn ] ) .
note that , the false vacuum energy density @xmath27|_{t = t_{\ast}}$ ] is the latent heat density , and the plasma thermal energy density @xmath28 is equal to @xmath29 . here
, @xmath30 is the temperature - dependent true minimum of the effective potential of the scalar field , which is responsible for the phase transition .
the parameter @xmath23 measures the strength of the phase transition gws , namely , a larger value for @xmath23 corresponds to a stronger first - order phase transition gws .
the second one is the time duration of the phase transition @xmath31 , where one has @xmath32 . in other words
, @xmath31 corresponds to the typical time scale of the phase transition .
since @xmath33 during the phase transition from its definition , one has @xmath34 the third one is the efficiency factor @xmath35 , which characterizes the fraction of the energy density converted into the motion of the colliding bubble walls and the last one is the bubble wall velocity @xmath36 .
the energy released into the gws of peak frequency is then given by @xcite : @xmath37 the second and third sources are the gws from the matter fluid effects , which can further contribute to the total energy released in gravitational radiation during the phase transition . here , we just use the formulae given in @xcite .
one source is from the sound waves in the fluid , where a certain fraction @xmath38 of the bubble wall energy ( after the collision ) is converted into motion of the fluid ( and is only later dissipated ) with the following contribution @xmath39 the other one is from turbulence in the fluid , where a certain fraction @xmath40 of the walls energy is converted into turbulence , with the contribution @xmath41 it is worth noticing that these two contributions from the matter fluid effects depend on @xmath42 linearly , and they are not fully understood . in some cases , these two effects may be larger than the one from bubble collisions .
the peak frequency produced by bubble wall collisions at @xmath26 during the first - order order phase transition is given by @xcite : @xmath43 .
considering the adiabatic expansion of our universe from the early universe to the present universe , the ratio of scale factors at the time of first - order phase transition and today can be written as @xmath44 where @xmath45 is the effective degrees of freedom at @xmath26 .
thus , the peak frequency today is @xmath46 , and the corresponding gw intensity is given by @xcite @xmath47 . \nonumber\end{aligned}\ ] ] the peak frequency of the gws signals due to the sound wave effects is about @xmath48 @xcite , and similarly its current value is @xmath49 . in this case , the gws intensity is expressed as @xcite @xmath50^{7/2 } , \nonumber\end{aligned}\ ] ] in which @xmath51 for relativistic bubbles @xcite .
the gws signals produced from the turbulence in the plasma have the peak frequency at about @xmath52 @xcite , which determines the current value as @xmath53 after considering the red shift .
the phase transition gws intensity from turbulence is formulated by @xcite @xmath54 the final phase transition spectrum consists of the three contributions above .
in this section , we discuss the phase transition gws in some non - abelian gauge group extended models in details , where one or several times symmetry breaking and first - order phase transition may occur with the evolution of our universe at certain critical temperature .
firstly , the gws spectra in the hidden gauge group extended models based on the @xmath2 gauge symmetry , commonly known as the 3 - 3 - 1 model @xcite are investigated .
the 3 - 3 - 1 models can explain the electric charge quantization and three generation @xcite .
the collider phenomenology of the 3 - 3 - 1 models has been extensively studied , such as the recent ref .
@xcite and references therein , and the phase transitions in some versions of 3 - 3 - 1 models have been studied by ref .
@xcite . here
, we try to explore the possible phase transition patterns associated with the symmetry breaking @xmath55 using the phase transition gws in some versions of the 3 - 3 - 1 models , where the scalars fields are accommodated in a certain representation of the @xmath56 gauge group in each version . in the following
, we will show three versions of 3 - 3 - 1 models ( the minimal , the economical and the reduced minimal 3 - 3 - 1 models , respectively ) @xcite , where very strong first - order phase transitions can take place at tev scale @xcite and produce detectable phase transition gws at elisa and bbo . for simplicity , we limit our discussion of the first - order phase transition to the thermal barrier case , where the potential barrier in the finite temperature effective potential origins from thermal loop effects . in this type of thermal barrier case ,
the bosonic fields contribute to the thermal effective potential of the form @xmath57 in the limit of high - temperature expansion . to begin with the concrete prediction of the gws in the following examples , we first discuss the general effective potential near the symmetry breaking and phase transition temperature , which can be approximated by @xmath58 here , x represents the field of the order parameter for the phase transition . for the electroweak phase transition in the sm ,
x field is just the higgs field .
the parameter @xmath59 quantify the interactions between @xmath60 field and the light bosons , and can be schematically written as @xmath61 . and , the parameters @xmath62 depends on interaction between @xmath60 and light bosons and fermions .
for the heavy fields whose mass is much larger than the critical temperature , their contribution can be omitted .
this can helps to simplify our discussions when the models have many new fields at different energy scales .
thus , in this case of qualitative analysis , the wash out parameter can be obtained as @xmath63 where the angle bracket means the vev of the field @xmath60 at the critical temperature @xmath64 . from the above qualitative analysis , we know that introducing new light bosonic fields ( compared to the corresponding critical temperature ) helps to produce or enhance the first - order phase transition .
the 3 - 3 - 1 models just introduce enough bosonic fields to produce detectable phase transition gws .
we consider the first - order phase transition gws spectrum in the so - called economical 3 - 3 - 1 model @xcite . in this version of 3 - 3 - 1 model , one chooses the simplest @xmath56 representations for the scalar fields with spontaneously symmetry breaking , namely , two complex scalar triplets with different y hypercharge are needed as follows @xmath65 @xmath66 the scalar potential is written as @xmath67 the @xmath68 gauge group is broken spontaneously via two steps . in the first step , symmetry breaking @xmath69 happens when the scalar triplet @xmath70 acquired with vacuum expectation value ( vev ) given by @xmath71 in the last step , to break into the sm @xmath72 gauge group @xmath73 , another scalar triplet @xmath74 is needed to acquire the vev as follows @xmath75 in this model , there exists two neutral scalars , one is the sm higgs boson @xmath76 , the other is the heavy scalar @xmath77 . in this paper , the packet ` cosmotransitions ' @xcite is used to numerically calculate the first - order phase transition . during the first symmetry breaking @xmath78 , the order parameter field for the phase transition is approximately the @xmath77 scalar field , namely , the @xmath79 field if it is compared to eq.([vappro ] ) .
strong first - order phase transition at the energy scale of several tev can be induced by the new bosons and exotic quarks in this 3 - 3 - 1 model if the mass of these new particle are from @xmath80 gev to several tev . during the last symmetry breaking @xmath73 ,
the order parameter field for the phase transition is approximately the higgs field , namely , the @xmath81 field if it is compared to eq.([vappro ] ) .
first - order phase transition at the electroweak scale can be trigger by the new bosons and it is rather weak than the first step . for the benchmark points of @xmath82 tev , @xmath83 tev , using the methods and formulae in the above section [ fgw ] , these two times first - order phase transition will produce two copies gws spectra with different characteristic peak frequency and amplitude , as shown in fig .
it shows that the phase transition gws can be used to explore the symmetry breaking and phase transition patterns at elisa and bbo . for the economical 3 - 3 - 1 model .
the colored regions correspond to the expected sensitivities of gw interferometers elisa and bbo , respectively .
the black line depicts the gw spectra for the two times first - order phase transition at @xmath84 at tev ( right line ) scale and @xmath85 at electroweak scale ( left line ) , respectively . ]
the gws spectra of the reduced minimal 3 - 3 - 1 model is similar to the one of the economical 3 - 3 - 1 model since their symmetry breaking and phase transition patterns are similar to the economical model .
the reduced minimal 3 - 3 - 1 model is mainly composed by neutral scalars @xmath76 , @xmath77 , doubly charged scalar @xmath86 , two sm like bosons @xmath87 , @xmath88 , the new heavy neutral boson @xmath89 , the singly and doubly charged boson @xmath90 and @xmath90 . these new particles and exotic quarks
can be triggers for the first - order phase transition @xcite .
taking use of the methods and formulae in sec.[fgw ] and the packet ` cosmotransitions ' @xcite , for the set of benchmark points @xmath91 tev , @xmath92 tev , two copies gws spectra during two times first - order phase transition will be produced , as shown in fig .
[ rm ] , which can also be detected by elisa and bbo .
the minimal 3 - 3 - 1 model @xcite corresponds to the parameter @xmath93
@xcite in the definition of the electric charge operator .
the gauge bosons , associated with the gauge symmetry @xmath94 of the model , consist of an octet @xmath95 ( @xmath96 ) and a singlet @xmath97 . in this version of 3 - 3 - 1 model ,
three @xmath56 triplets scalars are needed to break the symmetry and generate the mass of gauge bosons and exotic quarks as the following @xmath98 the scalar potential for@xmath99 , @xmath100 and @xmath70 is @xcite @xmath101 \left(\eta^\dagger\eta\right ) \nonumber \\ & + & \lambda_6\left(\rho^\dagger\rho\right)\left(\chi^\dagger\chi\right ) + \lambda_7\left(\rho^\dagger\eta\right)\left(\eta^\dagger\rho\right ) \nonumber \\ & + & \lambda_8\left(\chi^\dagger\eta\right)\left(\eta^\dagger\chi\right ) + \lambda_9\left(\rho^\dagger\chi\right)\left(\chi^\dagger\rho\right ) \nonumber \\ & + & \frac{1}{2 } \la f_1\epsilon^{ijk}\eta_i\rho_j\chi_k + { \mbox{h . c.}}\rf . \label{331m}\end{aligned}\ ] ] the new gauge fields acquire mass at several tev scale when the @xmath102 group breaks down to @xmath103 by the @xmath56 scalar triplet @xmath70 , while the ordinary quarks and sm gauge bosons obtain their masses at the last step symmetry breaking trigged by the triplets @xmath100 and @xmath99 .
there exist three neutral scalars include the lightest one which corresponds to the sm higgs boson @xmath76 and the other heavier scalar bosons @xmath104 and @xmath105 . by calculating the one - loop effective potential based on eq.([veff ] ) and taking use of the packet ` cosmotransitions ' @xcite in this minimal 3 - 3 - 1 model
, we find that there are regions of parameters allowed by the collider constraints that can favor a strong first - order phase transition , when the symmetry group spontaneously breaks from @xmath106 to @xmath107 as shown in ref .
@xcite . during this phase transition
, the x field here is just the @xmath104 field .
then , the phase transition gws spectrum can be obtained from the above gws spectrum formulae .
since this model has so many free parameters , which makes it very complicated to study the whole parameter regions allowed , we only show one benchmark point .
for example , when @xmath108 tev and @xmath109 tev , the corresponding gws spectrum is shown in fig .
[ min ] , which is within the sensitivities of elisa and bbo . for the reduced minimal 3 - 3 - 1 model .
the colored regions correspond to the expected sensitivities of gw interferometers elisa and bbo , respectively .
the black line depicts the gw spectrum for the two times first - order phase transition at @xmath84 at tev ( right ) and @xmath85 at electroweak scale ( left ) , respectively . ] for the minimal 3 - 3 - 1 model .
the colored regions correspond to the expected sensitivities of gw interferometers elisa and bbo , respectively .
the black line depicts the gw spectrum for the first - order phase transition at @xmath84 at tev scale . ] in general , if the sm is extended by hidden non - abelian gauge group , strong first - order phase transition may occur associated with the symmetry breaking at each symmetry breaking scale , where the phase transition gws may be produced . then , phase transition gws can be used to test the hidden gauge symmetry breaking .
besides the above phase transition gws in the 3 - 3 - 1 models , we first discuss new mechanism , namely , the so called relaxion mechanism @xcite proposed in 2015 , where the light higgs masss comes from the dynamical cosmological evolution during the early universe .
the relaxion mechanism can technically relax the ew hierarch @xcite and the highest cutoff relaxed in ref .
@xcite is about @xmath110 gev . the original relaxion mechanism in ref .
@xcite need the inflation sector and the qcd - like sector .
especially , one solution to avoid the strong cp problem in the simplest relaxion model , the dark qcd gauge group is needed , where new dark fermions are also included . in some allowed parameter space , the dark @xmath111 can give first - order qcd phase transition at the dark qcd scale .
if the dark qcd scale is about @xmath112 mev , the first - order phase transition can produce phase transition gws with the peak frequency in the @xmath113 hz range , which can be probed by the pta gws experiments , such as the ska or fast . a schematic gw spectrum for the dark qcd phase transition is shown in fig .
very recently , a novel mechanism called @xmath114naturalness " @xcite has been proposed to solve the electroweak hierarchy problem by introducing @xmath114 copies of the standard model with varying values of the higgs boson mass parameter in a unique universe .
although , the special reheaton particle has been added to the @xmath115naturalness mechanism to suppress the baryogenesis from other copyies with @xmath116 gev , the first - order phase transition many still happen in some parameter spaces and produce detectable gws signals , which will be carefully discussed in our future work @xcite .
however , we show that if a first - order phase transition takes place at a critical temperature of @xmath117(@xmath118@xmath110 ) gev , such as some unified models , this could potentially produce detectable gw spectrum in the future aligo experiments , and provide us with a unique probe of the hidden gauge symmetry breaking at high scales . from first - order phase transition during the evolution of our universe .
the colored regions represent the expected sensitivities of gws detectors aligo , elisa , bbo , ska , respectively .
the red line depicts the possible gws spectrum where the first - order phase transition occurs at the energy scale of @xmath119 mev in some dark qcd models .
the black line represents the gws spectrum for the first - order phase transition at about 4 tev in some dark gauge group models , such as some 3 - 3 - 1 models .
the purple line corresponds to the gws spectrum when the first - order phase transition takes place at the energy scale of @xmath120 tev in some models with high scale symmetry breaking . ]
in fig . [ gw ] , the first - order phase transition gws spectra @xmath121 during the evolution of our universe are shown for a generic classes of hidden gauge group extended models at the different energy scales .
the colored regions represent the expected sensitivities of gws detectors ska , bbo , elisa , aligo , respectively .
the red line depicts the possible gws spectrum where the first - order phase transition occurs at the energy scale of @xmath119 mev in some hidden qcd models .
the gws produced in the dark qcd models can be detected by the pta experiments , such as the planned ska or the fast built in china .
the black line represents the gws spectrum for the first - order phase transition at about 4 tev in some dark gauge group models , such as some versions of 3 - 3 - 1 models .
we have shown that all the three versions of the 3 - 3 - 1 models discussed above could produce strong first - order phase transition gws at tev scale when the dark gauge symmetry @xmath68 breaks to @xmath122 .
especially , in the economical and reduced minimal 3 - 3 - 1 models two times first - order phase transition may take place , which will produce two gws spectra with different characteristic peak frequencies .
in general , the phase transition gws produced at the scale from @xmath119 gev to several tev can be tested at future laser interferometer gws detectors in space , such as the planned elisa , bbo , taiji and tianqin .
the purple line corresponds to the gws spectrum when the first - order phase transition takes place at the energy scale of several tev in some high scale models , which may be within the sensitivity of the future aligo and provide us with a unique detection of the hidden gauge symmetry breaking at high scales beyond particle the abilities of particle colliders . besides these schematic models ,
many other hidden gauge group extended models can induce one time or several times first - order phase transition at different energy scale as shown in fig.[sb ] , where the corresponding phase transition gws spectra can be produced and tested at the corresponding gws detectors .
this approach can compensate for the collider experiments for exploring the hidden gauge group extended models and provide a novel approach to explore the symmetry breaking or phase transition pattern .
further study will be explored in our future study .
we thank andrew j. long , lian - tao wang and nima arkani - hamed for helpful discussions and comments during the workshop on cepc physics .
fph and xz are supported in part by the nsfc ( grant nos .
11121092 , 11033005 , 11375202 ) and by the cas pilotb program .
fph is also supported by the china postdoctoral science foundation under grant no .
2016m590133 .
999 b. p. abbott
_ et al . _ [ ligo scientific and virgo collaborations ] ,
lett . * 116 * , no . 6 , 061102 ( 2016 ) [ arxiv:1602.03837 [ gr - qc ] ] .
f. p. huang , y. wan , d. g. wang , y. f. cai and x. zhang , phys .
d * 94 * , no . 4 , 041702 ( 2016 ) doi:10.1103/physrevd.94.041702 [ arxiv:1601.01640 [ hep - ph ] ] .
p. schwaller , phys .
lett . * 115 * , 181101 ( 2015 ) [ arxiv:1504.07263 [ hep - ph ] ] .
p. s. b. dev and a. mazumdar , phys .
d * 93 * , no .
10 , 104001 ( 2016 ) doi:10.1103/physrevd.93.104001 [ arxiv:1602.04203 [ hep - ph ] ] . j. jaeckel , v. v. khoze and m. spannowsky , phys . rev .
d * 94 * , no .
10 , 103519 ( 2016 ) doi:10.1103/physrevd.94.103519 [ arxiv:1602.03901 [ hep - ph ] ] . h. yu , b. m. gu , f. p. huang , y. q. wang , x. h. meng and y. x. liu , arxiv:1607.03388 [ gr - qc ] .
a. addazi , arxiv:1607.08057 [ hep - ph ] .
p. huang , a. j. long and l. t. wang , phys . rev .
d * 94 * , no . 7 , 075008 ( 2016 ) doi:10.1103/physrevd.94.075008 [ arxiv:1608.06619 [ hep - ph ] ] .
f. p. huang and c. s. li , phys .
d * 92 * , no . 7 , 075014 ( 2015 ) doi:10.1103/physrevd.92.075014 [ arxiv:1507.08168 [ hep - ph ] ] .
f. p. huang , p. h. gu , p. f. yin , z. h. yu and x. zhang , phys .
d * 93 * , no .
10 , 103515 ( 2016 ) doi:10.1103/physrevd.93.103515 [ arxiv:1511.03969 [ hep - ph ] ] .
e. witten , phys .
d * 30 * , 272 ( 1984 ) . c. j. hogan , phys .
b * 133 * , 172 ( 1983 ) ; c. j. hogan , mon . not .
astron . soc .
* 218 * , 629 ( 1986 ) .
m. s. turner and f. wilczek , phys . rev .
lett . * 65 * , 3080 ( 1990 ) .
m. kamionkowski , a. kosowsky and m. s. turner , phys .
d * 49 * , 2837 ( 1994 ) [ astro - ph/9310044 ] .
m. hindmarsh , s. j. huber , k. rummukainen and d. j. weir , phys .
lett . * 112 * , 041301 ( 2014 ) [ arxiv:1304.2433 [ hep - ph ] ] ; c. caprini , r. durrer and g. servant , jcap * 0912 * , 024 ( 2009 ) [ arxiv:0909.0622 [ astro-ph.co ] ] .
m. hindmarsh , s. j. huber , k. rummukainen and d. j. weir , phys .
d * 92 * , 123009 ( 2015 ) [ arxiv:1504.03291 [ astro-ph.co ] ] .
f. pisano and v. pleitez , phys .
d * 46 * , 410 ( 1992 ) doi:10.1103/physrevd.46.410 [ hep - ph/9206242 ] .
p. h. frampton , phys .
lett . * 69 * , 2889 ( 1992 ) .
doi:10.1103/physrevlett.69.2889 p.
a. seoane _ et al .
_ [ elisa collaboration ] , arxiv:1305.5720 [ astro-ph.co ] .
v. corbin and n. j. cornish , class .
* 23 * , 2435 ( 2006 ) [ gr - qc/0512039 ] .
p. w. graham , d. e. kaplan and s. rajendran , phys .
lett . * 115 * , no . 22 , 221801 ( 2015 ) doi:10.1103/physrevlett.115.221801 [ arxiv:1504.07551 [ hep - ph ] ] .
r. smits , m. kramer , b. stappers , d. r. lorimer , j. cordes and a. faulkner , astron .
astrophys . * 493 * , 1161 ( 2009 ) doi:10.1051/0004 - 6361:200810383 [ arxiv:0811.0211 [ astro - ph ] ] .
g. hobbs , s. dai , r. n. manchester , r. m. shannon , m. kerr , k. j. lee and r. xu , arxiv:1407.0435 [ astro-ph.im ] . a. d. linde , nucl .
b * 216 * , 421 ( 1983 ) erratum : [ nucl .
b * 223 * , 544 ( 1983 ) ] .
doi:10.1016/0550 - 3213(83)90293 - 6 , 10.1016/0550 - 3213(83)90072-x s. r. coleman , phys .
d * 15 * , 2929 ( 1977 ) erratum : [ phys .
d * 16 * , 1248 ( 1977 ) ] .
doi:10.1103/physrevd.15.2929 , 10.1103/physrevd.16.1248 c. g. callan , jr . and
s. r. coleman , phys .
d * 16 * , 1762 ( 1977 ) .
doi:10.1103/physrevd.16.1762 c. grojean and g. servant , phys .
d * 75 * , 043507 ( 2007 ) [ hep - ph/0607107 ] ; c. caprini _ et al .
_ , jcap * 1604 * , no .
04 , 001 ( 2016 ) doi:10.1088/1475 - 7516/2016/04/001 [ arxiv:1512.06239 [ astro-ph.co ] ] . j. r. espinosa , t. konstandin , j. m. no and g. servant , jcap * 1006 * , 028 ( 2010 ) [ arxiv:1004.4187 [ hep - ph ] ] .
p. binetruy , a. bohe , c. caprini and j. f. dufaux , jcap * 1206 * , 027 ( 2012 ) [ arxiv:1201.0983 [ gr - qc ] ] .
s. j. huber and t. konstandin , jcap * 0809 * , 022 ( 2008 ) [ arxiv:0806.1828 [ hep - ph ] ]
. q. h. cao and d. m. zhang , arxiv:1611.09337 [ hep - ph ] .
j. s. borges and r. o. ramos , eur .
j. c * 76 * , no .
6 , 344 ( 2016 ) doi:10.1140/epjc / s10052 - 016 - 4168 - 8 [ arxiv:1602.08165 [ hep - ph ] ] .
v. q. phong , h. n. long , v. t. van and l. h. minh , eur .
j. c * 75 * , no . 7 , 342 ( 2015 ) doi:10.1140/epjc / s10052 - 015 - 3550 - 2 [ arxiv:1409.0750 [ hep - ph ] ] .
v. q. phong , v. t. van and h. n. long , phys .
d * 88 * , 096009 ( 2013 ) doi:10.1103/physrevd.88.096009 [ arxiv:1309.0355 [ hep - ph ] ] . c. l. wainwright , comput .
phys . commun .
* 183 * , 2006 ( 2012 ) doi:10.1016/j.cpc.2012.04.004 [ arxiv:1109.4189 [ hep - ph ] ] . c. salazar , r. h. benavides , w. a. ponce and e. rojas , jhep * 1507 * , 096 ( 2015 ) .
m. d. tonasse , phys .
b * 381 * , 191 ( 1996 ) doi:10.1016/0370 - 2693(96)00481 - 9 [ hep - ph/9605230 ] . | motivated by the discovery of gravitational waves ( gws ) and no evidence of new physics at current lhc , we discuss that many extended new physics models with hidden gauge group could undergo one or several times first - order phase transitions associated with the gauge group symmetry breaking in the evolution history of universe , which might produce detectable phase transition gws signals at future gws experiments , such as elisa . |
the higgs sector is ( so far ) an elusive basic ingredient of the fundamental theory of particle interactions .
searches for higgs bosons , and the study of their properties , are one of the primary goals of the large hadron collider ( lhc ) at cern , which is scheduled to commence taking data in the year 2007 .
the conventional folklore is that ( under reasonable model assumptions ) at least one higgs boson should be discovered at the lhc . in particular ,
if the light higgs predicted by the standard model ( sm ) exists , it will almost certainly be found at the lhc in the first years of running or even maybe before , at the tevatron .
moreover the lhc should provide a complete coverage of the sm higgs mass range .
however there is a strong belief that the standard model , in its minimal form with a single higgs , can not be the fundamental theory of particle interactions .
various extended models predict a large diversity of higgs - like bosons with different masses , couplings and even cp - parities .
the most elaborated extension of the standard model is , currently , the minimal supersymmetric standard model ( mssm ) in which there are three neutral ( @xmath7 , @xmath8 and @xmath9 ) and two charged ( @xmath10 ) higgs bosons , where @xmath7 and @xmath8 are cp - even ( @xmath11 ) and @xmath9 is cp - odd . just as for the standard model , this benchmark susy model has been studied in great detail ; for a recent review see @xcite . in the mssm ,
the properties of the higgs sector are characterized by the values of two independent input parameters , typically chosen to be the pseudoscalar higgs boson mass , @xmath12 , and the ratio , @xmath13 , of the vacuum - expectation - values of the two higgs doublet fields . at tree level ,
the pseudoscalar @xmath9 does not couple to the gauge bosons and its couplings to down- ( up- ) type fermions are ( inversely ) proportional to @xmath13 . within the mssm ,
the mass of the @xmath7-boson is bounded to @xmath14 gev ( see , for example , @xcite and references therein ) , while the experimental 95% cl lower limit for a neutral scalar higgs is @xmath15 gev @xcite . beyond the standard model the properties of the neutral higgs bosons can differ drastically from sm expectations . in some extended models , the limit for a neutral higgs can go down to below 60 gev .
this occurs , for instance in the mssm with explicit cp - violation , where the mass eigenstates of the neutral higgs bosons do not match the cp eigenstates @xmath16 , see , for example , @xcite .
further examples are the models with extra dimensions where a higgs can mix with the graviscalar of the randall sundrum scenario @xcite , which is called the radion ( see for example @xcite ) . in the latter case , due to the trace anomaly , the gluon gluon coupling is strongly enhanced , which makes this graviscalar especially attractive for searches in gluon - mediated processes , assuming that the background issues can be overcome .
these extended scenarios would complicate the study of the neutral higgs sector using the conventional ( semi)inclusive strategies . after the discovery of a higgs candidate the immediate task will be to establish its quantum numbers , to verify the higgs interpretation of the signal , and to make precision measurements of its properties .
the separation and identification of different higgs - like states will be especially challenging .
it will be an even more delicate goal to establish the nature of a newly - discovered heavy resonance state .
for example , how can one discriminate between the higgs of the extended susy sector from the graviscalar of the randall sundrum scenario ( or , even worse , from a mixture of the higgs and the radion ) .
as was shown in @xcite , the central exclusive diffractive processes ( cedp ) at the lhc can play a crucial role in solving these problems , which are so vital for the higgs physics .
these processes are of the form @xmath17 where the @xmath18 signs denote the rapidity gaps on either side of the higgs - like state @xmath19 .
they have unique advantages as compared to the traditional non - diffractive approaches @xcite .
these processes allow the properties of @xmath19 to be studied in an environment in which there are no secondaries from the underlying events .
in particular , if the forward protons are tagged , then the mass of the produced central system @xmath19 can be measured to high accuracy by the missing mass method . indeed , by observing the forward protons , as well as the @xmath20 pairs in the central detector
, one can match two simultaneous measurements of the @xmath19 mass : @xmath21 and @xmath22 .
moreover , proton taggers allow the damaging effects of multiple interactions per bunch crossing ( pile - up ) to be suppressed and hence offer the possibility of studying cedp processes at higher luminosities @xcite .
thus the prospects of the precise mass determination of the higgs - like states , and even of the direct measurements of their widths and @xmath20 couplings , looks feasible using these processes .
a promising option is to use the forward proton taggers as a spin - parity analyser @xcite
. this may provide valuable additional ( and in some cases unique ) information in the unambiguous identification of a newly discovered state . in section [ sec : disoverypotential ]
we illustrate the advantages of cedp , ( [ eq : cedp ] ) , in exploring the higgs sector in specific parameter ranges of the mssm .
first in section [ sec : icr ] we consider the so - called `` intense - coupling '' regime @xcite where the @xmath23 , @xmath24 and @xmath25 higgs decay modes are suppressed but where , on the other hand , the cedp cross section is enhanced , in comparison with the sm . next ,
in section 2.2 , we discuss the decoupling limit ( @xmath26 , @xmath27 ) where the light scalar , @xmath7 , looks very similar to the sm higgs . in this case
the discovery of the heavier scalar @xmath8 is crucial to establish the underlying structure of the electroweak - symmetry - breaking dynamics .
if @xmath28 gev , then the @xmath8 boson may be observed and studied in cedp . in section 2.3
we show that cedp may cover the `` windows '' where , once the @xmath7 boson is discovered , it is not possible to disentangle the @xmath8 or @xmath9 bosons by traditional means . in section 3
we explain how studies of cedp may help to determine the properties of the higgs - like states , once they have been discovered .
the plan of the paper is first to present the predictions and then , in section 4 , to explain how the cross sections for the cedp of the @xmath19 cp - even and cp - odd higgs states are calculated . in this section
we also enumerate the uncertainties in the predictions .
simple formulae to approximate the cedp cross sections for @xmath29 with @xmath20 , and the corresponding qcd @xmath30 background , are given in section 5 .
these allow quick , reliable estimates to be made of the signal - to - background ratio for a higgs of any mass , and both cp parities .
in section 6 we focus attention on ways to identify the pseudoscalar boson @xmath9 .
it is unlikely that the pure exclusive diffractive production process will have sufficient events , and so possible semi - inclusive signals are proposed and studied .
the main aim of this paper is to show how the higgs discovery potential at the lhc may be enlarged by using the unique advantages of forward proton tagging .
special attention is paid to the cases where the partial widths for the decay of higgs bosons into two gluons greatly exceed the sm values . as mentioned above
, we shall illustrate the advantages of using forward proton tagging to study the higgs sector by giving predictions for a few examples within the mssm model .
the `` intense - coupling regime '' @xcite is where the masses of all three neutral higgs bosons are close to each other and the value of @xmath13 is large .
then the @xmath31 decay modes ( which are among the main detection modes for the sm higgs ) are strongly suppressed .
this is the regime where the variations of all mssm higgs masses and couplings are very rapid .
this region is considered as one of the most challenging for the ( conventional ) higgs searches at the lhc , see @xcite and references therein . on the other hand , here
the cedp cross sections are enhanced by more than an order of magnitude ( due to the large @xmath32 couplings ) .
therefore the expected significance of the cedp signal becomes quite large .
indeed , this is evident from fig . 1 , which shows the cross sections for the cedp production of @xmath16 bosons as functions of their mass for @xmath33 and 50 .
the cross sections are computed as described in section 5 , with the widths and properties of the higgs scalar ( @xmath34 ) and pseudoscalar ( @xmath9 ) bosons obtained from the hdecay code , version 3.0 @xcite , with all other parameters taken from table 2 of @xcite ; also we take @xmath35 , which means the radiative corrections are included according to ref .
@xcite .
let us focus on the main @xmath36 decay mode decay mode .
the use of the @xmath37 decay mode requires an evaluation of the @xmath38 background , especially of the possibility of misidentifying gluon jets as @xmath39 s in the cedp environment .
currently , for inclusive processes , the probability , @xmath40 , to misidentify a gluon as a @xmath39 is estimated to be about 0.002 @xcite . ] .
this mode is well suited for cedp studies , since a @xmath41-even , @xmath42 selection rule @xcite suppresses the qcd @xmath30 background at lo .
indeed , at the lhc , the signal - to - background ratio is @xmath43 for a standard model higgs boson of mass @xmath44 gev , if the experimental cuts and efficiencies quoted in @xcite are used .
this favourable ratio was obtained since it was argued that proton taggers can achieve a missing mass resolution of @xmath45 gev . due to the enhancement of the production of mssm scalars , for large @xmath13
, the qcd background becomes practically negligible .
note that , normally , to estimate the statistical significance of the measurement of the signal cross section , we use the conservative formula @xmath46 for the statistical error .
however when we are looking for a new particle , which gives some peak on the top of the background , the statistical error may be much lower .
it is given just by the fluctuation of the background
@xmath47 . on the other hand for a low number of events ,
it is appropriate to use a poisson , rather than a gaussian , distribution .
thus the statistical significance to discover a new particle has to be evaluated more precisely . for instance , with the cuts and efficiencies quoted in [ 11 ] , we estimate that the qcd background is @xmath48 = 40(4 ) events for an integrated luminosity of @xmath49 ( 30 fb@xmath50 ) , if we take @xmath51 gev and @xmath52 gev .
thus , to obtain a confidence coefficient of @xmath53 ( equivalent to a statistical significance of @xmath54 for a gaussian distribution@xcite ) , it is sufficient for the cross section of a higgs signal , times the @xmath30 branching fraction , to satisfy @xmath55 if we take the integrated luminosity to be @xmath49 ( 30 fb@xmath50 ) . as can be seen from fig . 1 , for mssm with @xmath56 the expected cedp cross section times branching fraction , @xmath57 , is greater than 0.7fb for masses up to @xmath58 gev . the situation is worse for pseudoscalar , @xmath9 , production , since the cross section is suppressed by the @xmath41-even selection rule .
thus the cedp filters out pseudoscalar production , which allows the possibility to study pure @xmath8 production , see fig . 1
. this may be also useful in the decoupling limit , to which we now turn .
for @xmath59 and @xmath60 , there is a wide domain in mssm parameter space where the light scalar @xmath7 becomes indistinguishable from the sm higgs , and the other two neutral higgs states are approximately degenerate in mass .
this is the so - called decoupling limit @xcite .
moreover in many theories with a non - minimal higgs sector , there are significant regions of parameter space that approximate the decoupling limit ; see , for instance , @xcite . in this regime
the discovery of the heavier non - minimal higgs scalar @xmath8 is crucial for establishing the underlying structure of the electroweak - symmetry - breaking dynamics . here , forward proton tagging can play an important role in searching for ( at least ) the @xmath8-boson , if it is not too heavy ( @xmath61 gev ) . for large values of @xmath13
the decoupling regime essentially starts at @xmath62 gev . as can be seen in fig . 1
, the cross section is still sufficiently large to ensure the observation of the @xmath8 boson up to @xmath63 gev . to illustrate the moderate @xmath13 region we consider the case @xmath64 gev and @xmath65 . then the cedp cross section is @xmath66 thus for an integrated luminosity of @xmath49 we have about 8.5 events , after applying the experimental cuts and efficiencies discussed in ref . @xcite .
the corresponding background is about 2.6 events . using again a poisson distribution , as we did in the case of ( [ eq : cc1 ] ) , then the expected statistical significance to observe such an @xmath8 boson is about @xmath67 .
so detection may be feasible .
the possibility to use exclusive diffractive processes to explore larger masses will depend on various experiment - related factors .
in particular , on the prospects to achieve better mass resolution , @xmath68 , at higher mass , @xmath69 .
there are regions of mssm parameter space where , after the @xmath7 boson is discovered , it will be difficult to identify the heavy @xmath8 or @xmath9 bosons by conventional non - diffractive processes .
can the cedp processes help here ?
first consider the ` notorious hole ' in parameter space ( sometimes called the `` lhc wedge '' ) , defined by @xmath70 gev and @xmath7147 @xcite . in this region
, only the lightest higgs @xmath7 can be discovered even with integrated luminosity of the order of 300 fb@xmath50 .
its properties are nearly indistinguishable from those of a sm higgs .
unfortunately here , even for @xmath8 , the predicted cedp cross section is too low , assuming the efficiencies and cuts of ref .
@xcite ; see the heavy continuous curve in fig . 2
. however , with the present understanding of the capabilities of the detectors it is plausible that similar windows will occur in the interval @xmath72 gev for @xmath7346 .
that is , it may be impossible to identify the @xmath8 scalar by conventional processes at the @xmath54 confidence level with 300 fb@xmath50 of combined atlas+cms luminosity . from the predictions shown by the dashed line in fig . 2
, we see that the part of this ` hole ' , up to @xmath74 gev , may be covered by cedp . for illustration , consider the case @xmath75 gev and @xmath76 . with an integrated luminosity of @xmath77 , the cedp cross section of ( h)_h = 0.8 fb gives about 40 events , after applying the experimental cuts and efficiencies discussed in ref .
the corresponding background is about 10 events .
thus , if we use a poisson distribution , as we did in the case of ( [ eq : cc1 ] ) , then the expected statistical significance to observe such an @xmath8 boson will exceed @xmath78 .
recently some interesting applications of the cedp to studies of the more exotic higgs scenarios , such as models with cp violation and searches for the radions , have been discussed @xcite .
however the feasibility of applying these studies will require further critical discussion of the background issues .
as was discussed in refs .
@xcite , the use of cedp , with forward proton tagging , can provide a competitive way to detect a higgs boson .
this approach appears to have special advantages for searches of the cp - even higgses of mssm in the intense - coupling regime .
indeed , the cedp cross section can exceed the sm result by more than an order of magnitude , see fig . 1 .
once a higgs - like signal is observed , it will be extremely important to verify that it is indeed a higgs boson .
here we describe where forward proton tagging can provide valuable help to establish the nature of the newly discovered state and to verify whether it can be interpreted as a higgs boson .
we also emphasize that cedp are well suited to study specific cases which pose difficulties for probing the higgs sector at the lhc .
it is generally believed that a detailed coverage of the higgs sector will require complementary research programmes at the lhc and a linear @xmath79 collider .
interestingly , in some sense , the implementation of forward proton taggers allows some studies which otherwise would have to await the construction of a linear @xmath80 collider .
moreover , in addition to the other advantages , forward proton tagging offers the possibility to study new physics in the photon
photon collisions , see for example @xcite .
as mentioned above , if a candidate higgs signal is detected it will be a challenging task to prove its higgs identity .
unlike the conventional inclusive approaches , the very fact of seeing the resonance state in cedp automatically implies that the newly discovered state has the following fundamental properties .
it must have zero electric charge and be a colour singlet .
furthermore , assuming @xmath41 and @xmath81 conservation , the dominantly produced state has a positive natural parity , @xmath82 and even cp . recall that the installation of forward proton taggers also allows the attractive possibility of a spin - parity analysis .
this may provide valuable additional ( and in some cases unique ) leverage in establishing the origin of the discovered candidate state .
in particular , assuming cp conservation , the cedp allow the @xmath83 states to be filtered out , leaving only an ambiguity between the @xmath84 and @xmath85 states .
though without further efforts the @xmath85 state can not be ruled out , this would not be the most likely choice . as discussed in @xcite , studying of the azimuthal correlations of the outgoing protons can allow further spin - parity analysis . in particular , it may be possible to isolate the @xmath86 state .
the azimuthal distribution distinguishes between the production of scalar and pseudoscalar particles @xcite .
note that with the forward protons we can determine the cp - properties of the higgs boson irrespective of the decay mode .
moreover , cedp allow the observation of the interference effects between the cp - even and cp - odd @xmath87 transitions ( see @xcite ) . following the discovery , and the determination of the basic quantum numbers , of the new state , the next step is to check whether the coupling properties match those expected for the higgs boson .
for this , we need to detect the signal in at least two different channels .
the cedp mechanism provides an excellent opportunity background . ] to detect the @xmath30 mode .
the other decay channel which is appropriate for cedp studies is the @xmath88 decay mode ( or @xmath24 , or @xmath89 , if the event rate is sufficient ) .
these measurements allow a non - trivial check of the theoretical models .
for example , the precise measurement of the partial decay widths into tau and bottom quark pairs can provide a way to discriminate between supersymmetric and non - supersymmetric higgs sectors , and to probe cp - violating effects ( for recent papers see , for instance , @xcite and references therein ) . for illustration ,
consider the following topical example , namely the intense coupling limit of mssm .
as mentioned before , this regime is especially difficult for conventional inclusive studies of the mssm higgs @xcite . on the other hand , with forward proton taggers ,
cedp offer an ideal probe . for numerical purposes ,
we choose the same parameters as in @xcite , in particular , @xmath33 , together with results for @xmath56 .
the salient features of this regime are that there is almost a mass degeneracy of the neutral higgs states , and their total widths can be quite large ( due to the @xmath90 increase ) and reach up to 12 gev .
this can be seen from fig . 3 , where we plot the mass differences @xmath91 and @xmath92 , and the widths of the bosons , as a function of the pseudoscalar mass @xmath12 .
it is especially difficult to disentangle the higgs bosons in the region around @xmath93 gev , where all three neutral higgs states practically overlap @xcite .
since the traditional non - diffractive approaches do not , with the exception of the @xmath23 and @xmath94 modes , provide a mass resolution better than 1020 gev , all three higgs bosons will appear as one resonance decay mode is hopeless .
moreover , the vector - boson - fusion higgs production , with the @xmath30 or @xmath37 decays , does not allow the separation of the @xmath7 and @xmath8 peaks due to the poor invariant mass resolution .
as shown in @xcite , only the detection of the rare dimuon higgs decay mode may provide a chance to separate the higgs states in inclusive production .
even this would require that the higgs mass splitting exceeds at 35 gev . ] .
as an example , we give the expected number of cedp signal and background events for @xmath95 for @xmath96 gev , using the efficiencies and angular cuts adopted in ref . @xcite .
for @xmath13 = 30(50 ) , the number of the @xmath97 signal events , @xmath98 = 0.35(1 ) , while the qcd background is @xmath99 for @xmath52 gev . for this choice of parameters , @xmath100 gev , and
the number of the @xmath101 events is @xmath102 = 30(71 ) with background @xmath48 = 3.5(3.2 ) .
the corrresponding numbers for the @xmath8 boson are @xmath103 gev , @xmath104 and @xmath105 .
an immediate advantage of cedp , for studying this troublesome region , is that the @xmath9 contribution is strongly suppressed , while the @xmath7 and @xmath8 states can be well separated ( @xmath106 gev ) given the anticipated experimental mass resolution of @xmath107 gev @xcite , see fig .
note that precision measurements of the two cp - even higgs states may allow a direct determination of the basic mssm parameter , @xmath13 .
moreover , the forward tagging approach can provide a direct measurement of the width , becomes comparable to the mass resolution , @xmath68 , the zero - width approximation is not valid any more .
therefore in calculations of the cross sections the mass resolution and large - width effects should be taken into account .
] of the @xmath7 ( for @xmath108 gev ) and the @xmath8-boson ( for @xmath109 gev ) . outside the narrow range @xmath110 gev ,
the widths of the @xmath7 and @xmath8 are quite different ( one is always much narrower than the other ) , see fig .
3 . recall that the width measurement can play an important role in the understanding of higgs dynamics .
it would be instructive to observe this phenomenon experimentally .
as discussed in @xcite , for @xmath111 the central exclusive signal should be still accessible at the lhc up to an @xmath8 mass about 250 gev .
for instance , for @xmath112 gev , and lhc luminosity @xmath113 ( @xmath114 ) , about 20 ( 200 ) @xmath115 events are produced .
if the experimental cuts and efficiencies quoted in dkmor are imposed , then the signal is depleted by about a factor of 6 .
this leaves 3 ( 30 ) observable events , with background of about 0.1 ( 1 ) events .
to calculate the cross section for the central exclusive diffractive production of @xmath117 higgs bosons rather than @xmath118 .
] we use the formalism of refs .
the amplitudes are described by the diagram shown in fig .
[ fig:3](a ) , where the hard subprocesses @xmath119 are initiated by gluon
gluon fusion and where the second @xmath120-channel gluon is needed to screen the colour flow across the rapidity gap intervals . ignoring , for the moment , the screening corrections of fig .
[ fig:3](b ) , the born amplitudes are of the form @xcite t_h = a^2 f_g(x_1 , x_1 , q_1 ^ 2 , ^2 ; t_1)f_g(x_2,x_2,q_2 ^ 2,^2 ; t_2 ) , [ eq : rat3 ] where the @xmath119 subprocesses are specified by @xcite @xmath121 with the nlo @xmath122 factor @xmath123 @xcite , and by the factors l l l v_h(0^+ ) & = & ( q_- p_1 ) ( q_+ p_2 ) + v_h(0 ^ - ) & = & ( ( q_- p_1 ) ( q_+ p_2))n_0 .
[ eq : rat4 ] here @xmath124 is a unit vector in the beam direction . the @xmath125 s are the skewed unintegrated gluon densities of the proton at the hard scale @xmath126 , taken to be @xmath127 , with l l l q_1 & = & \{q_,|(q_- p_1)| } , + q_2 & = & \{q_,|(q_+ p_2)|}. [ eq : rat5 ] the longitudinal momentum fractions carried by the gluons satisfy @xmath128 where , for the lhc , with @xmath129 tev , we have @xmath130 , while @xmath131 .
below , we assume factorization of the unintegrated distributions , f_g(x , x,q^2,^2;t ) = f_g(x , x,q^2,^2)f_n(t ) , [ eq : rat5a ] where we parameterize the form factor of the proton vertex by the form @xmath132 with @xmath133 . in the domain specified by ( [ eq : lmf_ineq ] ) the skewed unintegrated densities are given in terms of the conventional ( integrated ) densities @xmath134 . to single log accuracy , we have @xcite .
f_g(x , x,q_i^2,^2 ) = r_g ( xg(x , q_i^2 ) ) , [ eq : rat6 ] where @xmath135 is the usual sudakov form factor which ensures that the gluon remains untouched in the evolution up to the hard scale @xmath126 , so that the rapidity gaps survive .
this sudakov factor @xmath135 is the result of resumming the virtual contributions in the dglap evolution .
it is given by @xmath136dz\right ) .
\label{eq : t}\ ] ] here we wish to go beyond the collinear approximation and in the @xmath135 factor to resum , not just the single collinear logarithms , but the single soft @xmath137 terms as well .
thus we consider the region of large - angle soft gluon emission , and explicitly calculate the one - loop vertex diagram .
in particular , we do not approximate @xmath138 by @xmath139 .
then we adjust the upper limit of the @xmath140 integration in ( [ eq : t ] ) to reproduce the complete one - loop result gives @xmath141dk^2_\perp / k^2_\perp$ ] .
the last number in the square brackets corresponds to the standard non - logarithmic contribution related to the @xmath142 term in ( [ eq : t ] ) , while the first number stems from the exact treatment of large - angle gluon radiation .
this result can be incorporated in ( [ eq : t ] ) by simply choosing @xmath143 as in ( [ eq : monday1 ] ) . ] .
we find = .
[ eq : monday1 ] the square root in ( [ eq : rat6 ] ) arises because the bremsstrahlung survival probability @xmath135 is only relevant to hard gluons .
@xmath144 is the ratio of the skewed @xmath145 integrated distribution to the conventional diagonal density @xmath146 .
for @xmath147 it is completely determined @xcite , with an uncertainty in @xmath144 of the order of @xmath148 .
the apparent infrared divergence of ( [ eq : rat3 ] ) is nullified domain @xcite . ] for @xmath149 production by the sudakov factors embodied in the gluon densities @xmath125 . however , as discussed in ref .
@xcite , the amplitude for @xmath150 production is much more sensitive to the infrared contribution .
indeed let us consider the case of small @xmath151 of the outgoing protons .
then , from ( [ eq : rat4 ] ) , we see that @xmath152 , whereas @xmath153 ( since the linear contribution in @xmath154 vanishes after the angular integration ) .
thus the @xmath155 integration for @xmath149 is replaced by @xmath156 for @xmath150 , and now the sudakov suppression is not enough to prevent a significant contribution from the @xmath157 domain . to estimate the uncertainty in the predictions for the @xmath158 exclusive diffractive cross sections we first quantify the above uncertainty arising from the infrared region , where the gluon distribution is not well known . as an example ,
consider @xmath150 and @xmath149 production at the lhc , for @xmath159 gev and @xmath160 , using different treatments of the infrared region .
we perform calculations using mrst99 @xcite and cteq6 m @xcite partons respectively with the very low @xmath161 gluon frozen at its value at @xmath162 gev
. then we integrate down in @xmath154 until @xmath163 are close to @xmath164 , where the contribution vanishes due to the presence of the @xmath135-factor .
this will slightly overestimate the cross sections , as for @xmath165 the gluon density has a positive anomalous dimension ( @xmath166 with @xmath167 ) and decreases with decreasing @xmath168 .
a lower extreme is to remove the contribution below @xmath162 gev entirely .
even with this extreme choice , the @xmath169 cross section is not changed greatly ; it is depleted by about 20% . on the other hand , as anticipated , for @xmath86 production , the infrared region is much more important and the cut reduces the cross section by a factor of 5 . another uncertainty is the choice of factorization scale @xmath126 .
note that in comparison with previous calculations @xcite , which were done in the limit of proton transverse momenta , @xmath170 , now we include the explicit @xmath171-dependence in the @xmath154-loop integral of ( [ eq : rat3 ] ) .
we resum the ` soft ' gluon logarithms , @xmath172 , in the @xmath135-factor .
so now the @xmath135-factor includes both the soft and collinear single logarithms .
the only uncertainty is the non - logarithmic nlo contribution
. this may be modelled by changing the factorization scale , @xmath126 , which fixes the maximal @xmath173 of the gluon in the nlo loop correction . as the default we have used @xmath160
; that is the largest @xmath173 allowed in the process with total energy @xmath174 .
choosing a lower scale @xmath175 would enlarge the cross sections by about 30% . recall that the general calculation is performed in the collinear approximation .
however , when we take account of large - angle soft gluon emission in the form factor , that is in the calculation of the virtual - loop correction , as in ( [ eq : t ] ) and ( [ eq : monday1 ] ) , we go beyond the collinear approximation .
the corresponding real contribution in ( [ eq : rat6 ] ) is represented by @xmath144 and the derivative of the @xmath176 factor . strictly speaking , for the case of large - angle emission there will be some small deviation from this simplified formula .
this gives an extra uncertainty of the order of 10% .
next there is some uncertainty in the gluon distribution itself . to evaluate this , we compare predictions obtained using cteq6 m @xcite , mrst99 @xcite and mrst02 @xcite partons . for @xmath169 production at the lhc , with @xmath159 gev and @xmath160
, we find that the effective gluon
gluon luminosity , before screening , is @xmath177 respectively .
this spread of values arises because the cteq gluon is 7% higher , and the mrst02 gluon 4% lower , than the default mrst99 gluon , in the relevant kinematic region .
the sensitivity to the gluon arises because the central exclusive diffractive cross section is proportional to the 4th power of the gluon . for @xmath86 production ,
the corresponding numbers are @xmath178 . up to now
, we have discussed the effective gluon gluon luminosity .
however , nnlo corrections may occur in the @xmath179 fusion vertex .
these give an extra uncertainty of @xmath180 .
note that we have already accounted for the nlo corrections for this vertex @xcite .
furthermore , we need to consider the soft rescattering which leads to a rather small probability , @xmath181 , that the rapidity gaps survive the soft @xmath182 interaction , see fig .
[ fig:3](b ) . from the analysis @xcite of all soft @xmath182 data we estimate the accuracy of the prediction for @xmath183 is @xmath184 .
one check of the eikonal model calculations of @xmath183 is the estimate of the diffractive dijet production rate measured by the cdf collaboration @xcite at the tevatron .
the rate , when calculated using factorization and the diffractive structure functions obtained from hera data , lies about a factor of 10 above the cdf data .
however , when rescattering corrections are included , and the survival probabilities computed , remarkably good agreement with the cdf measurements is obtained @xcite .
combining together all these sources of error we find that the prediction for the @xmath169 cross section is uncertain to a factor of almost 2.5 , that is up to almost 2.5 , and down almost to @xmath185 , times the default value .
gev , to be 2.2 fb with an uncertainty given by the range 0.95.5 fb . ] on the other hand , @xmath86 production is uncertain by this factor just from the first ( infrared ) source of error , with the remaining errors contributing almost another factor of 2.5 .
here we give simple approximate formulae for the cross sections for the central exclusive diffractive production of @xmath117 bosons , @xmath186 , and for the qcd @xmath30 background as a function of mass . as was shown in ref .
@xcite , the cross section may be written as the product of the effective gluon
gluon luminosity @xmath187 and the cross section for the hard subprocess @xmath188 _ cedp = l .
[ eq:6a]at the lhc energy , the luminosity , integrated over rapidity , may be approximated , to an accuracy of better than 10% in the interval @xmath189 gev , by = dy = \
{ c l & for h(0^+ ) production + + & for h(0 ^ - ) .
[ eq : integlumin ] where the mass @xmath69 is in gev .
the factors @xmath190 , which are the probability that the rapidity gaps survive the soft rescattering , are 0.026 and 0.087 for @xmath149 and @xmath150 respectively .
the @xmath86 amplitude contains the kinematic factor @xmath191 , which implies that in impact parameter space the fourier transform contains the factor @xmath192 @xcite .
thus the @xmath86 amplitude tends to populate larger impact parameters , where the suppression caused by soft rescattering is less effective , than does the @xmath169 amplitude . the @xmath179 subprocesses cross section is , in the zero width approximation , _
h = ( 1- ) ~_s^2(m_h^2)g_f , where @xmath193 is the fermi constant .
the @xmath30 background processes were discussed in detail in ref .
@xcite for a standard model @xmath194 signal , with @xmath195 gev .
there were four appreciable sources of @xmath30 background , each with a background - to - signal ratio in the range @xmath1960.08 , after the appropriate cuts . in brief , these give contributions to the background subprocess cross sections of the following approximate forms * @xmath197(@xmath30 mimicked by @xmath198 ) : @xmath199 , * @xmath200(@xmath201 admixture ) : @xmath202 , * @xmath203(@xmath204 , @xmath42 contribution ) : @xmath205 , * nlo @xmath206 contribution : @xmath207 , where the renormalisation scale is specified by @xmath208 .
the @xmath209 superscript is to denote that each incoming gluon to the hard subprocess belongs to a colour - singlet @xmath120-channel state , see fig .
[ fig:3 ] .
note that all background contributions go like @xmath210 , except for ( iii ) .
in addition , for fixed missing mass resolution @xmath68 , the background contains a factor @xmath211 .
thus for large @xmath212 gev the background cross section behaves roughly as _
b = l _ b ~m / m^6 , while the higgs signal behaves as _
h = l _ h ~1/m^3 . to obtain a more precise behaviour
one can use ( [ eq : integlumin ] ) . in ref .
@xcite we found a signal - to - background ratio @xmath213 for a @xmath195 gev standard model higgs boson .
the ratio may , however , be appreciably enhanced for susy higgs with large @xmath69 , and/or large @xmath13 .
it is interesting to note that , in the region where the background is larger than the signal ( @xmath214 ) , the statistical significance of the signal @xmath215 is practically independent of mass , assuming that the experimental resolution @xmath68 is mass independent .
moreover , for a low mass , where @xmath30 pairs in the @xmath42 state , ( iii ) , give the dominant contribution to the background , the statistical significance increases with mass , as here @xmath216 . finally , it is important to note that there are still various open questions which must be addressed by experimentalists in order to establish the feasibility of cedp for higgs searches at the lhc .
one of the most challenging issues concerns triggering .
assuming that the trigger is based on the central detector only , the main task is to reduce the rate of triggering jets in order not to reach saturation of the trigger .
we present below some numbers which may be useful when considering this issue . based on the results of ref .
@xcite , we see that the cross section for the cedp of dijets with @xmath217 gev is 200 pb . if we assume that the proton tagging efficiency is 0.6 , then this yields 0.12 events / sec for a luminosity of @xmath218 . at the same time , the inclusive cross section for production of a pair of jets , each with @xmath217 gev is about 50 @xmath126b .
this leads to 50,000 events / sec , which greatly exceeds the trigger saturation limit . for cms , for example
, this latter limit is 100 events / sec @xcite .
therefore it appears necessary to devise additional topological requirements in the level-1 trigger for the exclusive signal , such as making use of the presence of the rapidity gaps .
if the exclusive diffractive cross sections for scalar and pseudoscalar higgs production were comparable , it would be possible to separate them readily by the missing mass scan , and by the study of the azimuthal correlations between the outgoing proton momenta .
however , we have seen that the cross section for pseudoscalar higgs exclusive production is , unfortunately , strongly suppressed by the @xmath41-even selection rule , which holds for cedp at lo . for values of @xmath21915 ,
the separation between the @xmath220 and @xmath9 bosons is much larger than their widths , see fig .
. hence it might be just possible to observe the pseudoscalar in cedp .
for example , for @xmath221 and @xmath222 gev , the mass separation , 3.6 gev , between @xmath7 and @xmath9 is about 8 times larger than the width @xmath223 .
the cross section ( a)_a 0.15 fb , when allowing for the large uncertainties in @xmath224 , could be just sufficient to bring the process to the edge of observability .
on the other hand , one may consider the semi - exclusive process ppp + jj + p , [ eq : sep ] where the higgs bosons , @xmath225 , are accompanied by two gluon jets with transverse momenta @xmath171 satisfying @xmath226 .
the emission of soft gluons , with energies ( in the @xmath19 rest frame ) @xmath227 , does not affect the @xmath41-even selection rule .
thus in this region the scalar @xmath8 boson will be dominantly produced in process ( [ eq : sep ] ) .
on the other hand , in the domain x_i 2e_i / m_~1 we lose the selection rule and have comparable @xmath8 and @xmath9 production .
moreover the azimuthal angular distribution production was mentioned in ref .
however , the proof of the feasibility of such an approach in non - diffractive @xmath228 production requires further study of the possible dilution of the effect caused by parton showers ( see also @xcite ) .
an exclusive environment of the jets may be needed . ] for the gluon jets is different for the @xmath8 and @xmath9 cases .
in particular , for gluons with @xmath229 accompanying @xmath9 production , the azimuthal distribution is practically flat ( at lo ) , whereas for @xmath8 production it has the form @xmath230 .
this observation may be used to distinguish between @xmath8 and @xmath9 production , and to study better the properties of the @xmath9 boson .
of course , we have to pay a price , of about a factor @xmath231 , to produce two extra jets with @xmath171 in the range @xmath232 gev , in a rapidity interval with @xmath233 .
however , this is less than the suppression of @xmath9 production by the @xmath41-even selection rule . for @xmath234 and @xmath235 gev the cross section still exceeds 1 fb . to gain more insight , it is informative to return to the original central exclusive diffractive process , which we sketch in fig .
[ fig:3](a ) . in the equivalent gluon approximation ,
the polarization vectors @xmath236 of the gluons , carrying momentum fractions @xmath237 , are aligned along the corresponding gluon transverse momenta @xmath238 and @xmath239 .
for very forward protons with @xmath240 , this leads to a correlation between the polarization vectors , namely @xmath241 is parallel to @xmath242 .
on the other hand , the @xmath243 production vertices behave as @xmath244 and @xmath245 .
therefore the exclusive diffractive production of the pseudoscalar @xmath9 boson is strongly suppressed are parallel to @xmath246 , it follows , after integration over the azimuthal angle of @xmath246 , that the lo qcd production of @xmath30 pair is also suppressed by a factor @xmath247 ; which results in a very low qcd @xmath248 background to the cedp . ] . to change the @xmath120-channel gluon polarization
we need particle emission
. the best possibility would be to emit either a @xmath249 or a @xmath250 meson .
unfortunately the corresponding cross sections are too low .
we lose more than a factor of @xmath251 , which is a stronger suppression than that coming from the @xmath41-even , @xmath252 selection rule .
for this reason we considered the possibility of emitting a ( colour singlet ) pair of gluons , that is process ( [ eq : sep ] ) . for soft @xmath253 emission ,
the dominant contribution comes from the term in the triple - gluon vertex which contains @xmath254 , and so does not change the gluon polarization vector . here
the indices @xmath255 and @xmath256 correspond to the @xmath120-channel gluon polarization before and after the emission of the @xmath257-channel gluon .
however for large @xmath258 the major contribution comes from the term with @xmath259 , where @xmath260 is the @xmath257-channel gluon polarization index . in this case after the emission , the new @xmath120-channel polarization vector is directed along the new transverse momentum @xmath261 .
thus for @xmath262 the cross sections for @xmath8 and @xmath9 production are comparable , but now the lo qcd background is no longer suppressed .
maybe the best chance to identify the @xmath118 boson is to observe the double - diffractive inclusive process pp x + + y , [ eq : a1 ] where both protons are destroyed . in terms of @xmath120-channel gluon polarization , this is equivalent to the limit @xmath263 in the previous reaction ( [ eq : sep ] ) .
process ( [ eq : a1 ] ) has the advantage of a much larger cross section , which came from the @xmath264 integrals limited by the proton form factors , as in ( [ eq : rat5a ] ) ( see also eq.(8 ) in ref.@xcite ) , here we have logarithmic @xmath265 integrals over the intervals @xmath266 .
moreover , the sudakov supression ( which arises from the requirement that no extra relatively hard gluons are emitted with transverse momenta @xmath267 in the interval @xmath268 ) is much weaker than for the exclusive process due to the larger values of @xmath151 , see @xcite .
therefore the effective @xmath269 luminosity for double - diffractive inclusive production is much larger than in the cedp case . ] .
however , just as for the above semi - exclusive process , we do not have the @xmath252 selection rule to suppress the @xmath248 background , nor do we have the possibility of the good missing mass resolution . on the other hand , the @xmath269 luminosity is more than order of magnitude larger than for the pure exclusive case . for example , for double - diffractive inclusive production , with the rapidity gaps @xmath270 ,
the luminosity is 20 times larger than that for the exclusive diffractive production of a higgs boson with mass @xmath271 gev ( see for details @xcite ) .
so , even for the @xmath37 decay mode we expect a cross section , @xmath272 , of about 20 fb for @xmath9 production in the mssm with @xmath33 and @xmath273 gev .
this looks promising , provided that the probability for gluon misidentification as a @xmath39 is less than 1/150 , which appears to be feasible@xcite . despite the qcd background
, the observation of the @xmath9 boson also appears to be feasible for the @xmath248 decay mode , since then @xmath274 fb which gives a signal - to - background ratio of @xmath275 .
we have evaluated the qcd @xmath248 background using the lo formula for the @xmath276 cross section ( without any @xmath122 factor ) with the polar angle cut @xmath277 as was done in @xcite , and assumed a mass resolution of @xmath278 gev .
the efficiency of tagging the @xmath279 jets was taken to be about 0.6 @xcite .
this results in a statistical significance factor , @xmath280 , for the background then the statistical significance is about @xmath281 . ] of the signal for the @xmath9 boson of @xmath282 , for an integrated luminosity of 30 fb@xmath50 .
moreover one can measure the azimuthal angle @xmath283 between the transverse energy @xmath284 flows in the forward and backward regions , that is between the total transverse momenta of the @xmath285 and @xmath286 systems .
the angular dependence is driven by the momenta transferred across the rapidity gaps , and has the form & ^2 & h(0^+ ) + and & ^2 & a(0 ^ - ) , whereas for the qcd @xmath30 background the @xmath283 dependence is flat ( if we neglect the quark mass ( @xmath287 ) and average over the @xmath279-jet momentum direction ; see eq.(15 ) of @xcite for @xmath288 ) .
thus we may suppress the @xmath289 signal by selecting events with @xmath283 close to @xmath290 .
we emphasize that here we deal with rather large @xmath291 jets , @xmath292 gev , and so the angular dependences are largely insensitive to the soft rescatterings .
furthermore , instead of measuring the @xmath293 of the jets ( as in ref.@xcite ) , for process ( [ eq : a1 ] ) it is enough to measure the total transverse energy @xmath284 in the forward ( and backward ) regions of proton dissociation .
recall that the effective @xmath269 luminosity for the inclusive process ( [ eq : a1 ] ) was evaluated using the leading order formula .
thus we can not exclude rather large nlo corrections , particularly for the non - forward bfkl amplitude .
on the other hand , here , we do not enter the infrared domain .
moreover , with `` skewed '' kinematics , the nlo bfkl corrections are expected to be much smaller than in forward case .
so we may expect that the uncertainty of the prediction is about factor of 3 - 4 ( or even better ) .
the central exclusive diffractive processes promise a rich physics menu for studying the mechanism of electroweak symmetry breaking , and the detailed properties of the higgs sector .
we have seen , within mssm , that the expected cedp cross sections are large enough , especially for large @xmath13 , both in the intense coupling and the decoupling regimes .
thus cedp offers a way to cover those regions of mssm parameter space which may be hard to access with the conventional ( semi ) inclusive approaches .
this considerably extends the physics potential of the lhc and may provide studies which are complementary , both to the traditional non - diffractive approaches at the lhc , and to the physics program of a future linear @xmath79 collider .
to some extent this approach gives information which otherwise may have to await a linear @xmath80 collider .
we thank alfred bartl , albert de roeck , abdelhak djouadi , howie haber , leif lonnblad , walter majerotto , risto orava , andrei shuvaev , michael spira , and especially sasha nikitenko and georg weiglein , for useful discussions . adm
thanks the university of canterbury ( new zealand ) for an erskine fellowship and the leverhulme trust for an emeritus fellowship .
this work was supported by the uk particle physics and astronomy research council , by grants intas 00 - 00366 , rfbr 01 - 02 - 17383 and 01 - 02 - 17095 , and by the federal program of the russian ministry of industry , science and technology 40.052.1.1.1112 and ss-1124.2003.2 .
cms collaboration , technical proposal , report cern / lhcc/94 - 38(1994 ) ; + atlas collaboration , technical design report , cern / lhcc/99 - 15(1999 ) ; + d. denegri et al . , arxiv : hep - ph/0112045 ; + b. mellado , for cms and atlas collaborations , arxiv : hep - ex/0211062 .
khoze , a.d .
martin and m.g .
ryskin , hep - ph/0006005 , in _ proc . of 8th int .
workshop on deep inelastic scattering and qcd ( dis2000 ) _ , liverpool , eds .
j. gracey and t. greenshaw ( world scientific , 2001 ) , p.592 . | we show that forward proton tagging may significantly enlarge the potential of studying the higgs sector at the lhc .
we concentrate on higgs production via central exclusive diffractive processes ( cedp ) .
particular attention is paid to regions in the mssm parameter space where the partial width of the higgs boson decay into two gluons much exceeds the sm case ; here the cedp are found to have special advantages .
ippp/03/51 + dcpt/03/102 + 4 december 2003 + * extending the study of the higgs sector at the lhc by proton tagging * a.b .
kaidalov@xmath0 , v.a .
khoze@xmath1 , a.d . martin@xmath2 and m.g .
ryskin@xmath1 + @xmath3 department of physics and institute for particle physics phenomenology , + university of durham , dh1 3le , uk + @xmath4 institute of theoretical and experimental physics , moscow , 117259 , russia + @xmath5 petersburg nuclear physics institute , gatchina , st . petersburg , 188300 , russia + @xmath6 department of physics , university of canterbury , christchurch , new zealand + |
accretion disks are important in several astrophysical systems . they can be found around young stellar objects(yso ) , around compact stellar objects in our galaxy and around several super - massive black holes in active galactic nuclei(agn ) . in super massive accretion disks
the mass of black hole is @xmath0 . to study such disks we use the general relativity with relativistic hydrodynamics in kerr metric background geometry . in relativistic navier - stokes fluid
we have stress - energy tensor which is related to viscosity and is the cause of redistribution of energy and momentum in fluid .
this tensor is defined by the four velocity and metric . in the previous studies ,
the relativistic and stationary solutions of standard black hole disks were solved .
lasota ( 1994 ) was the first who wrote down slim - disk equations which include relativistic effects .
he also assumed that only r-@xmath1 component of the stress tensor is nonzero .
he used a special form for this component , which was followed by abramowicz et al(1996 and 1997 ) .
chakrabarti ( 1996 ) derived the transonic solutions of thick and thin disks for a weak viscosity .
he assumed similar form for viscosity as lasota ( 1994 ) .
then manmoto ( 2000 ) derived the global two temperatures structure of advection - dominated accretion flows ( adafs ) numerically by using full relativistic hydrodynamical equations including the energy equations for the ions and electrons .
papaloizou & szuskiewicz ( 1994 ) introduced a phenomenological and non - relativistic equation for the evolution of viscous stress tensor ( causal viscosity ) which has been used by many authors .
gammie & popham ( 1998 ) and popham & gammie ( 1998 ) solved adafs with relativistic causal viscosity .
they used the boyer - lindquist coordinates .
takahashi ( 2007b ) solved the equations of relativistic disks in the kerr - schild coordinates by using the relativistic causal viscosity . in the papers of gammie & popham and takahashi they assumed that in the frf
, only the @xmath2 component of shear viscosity is nonzero , then they used the transformation tensors to derive the components of shear stress tensor in the boyer - lindquist or kerr - schild frames .
in present study , we concentrate on the stationary axisymmetric accretion flow in the equatorial plane .
we use a new method to calculate the shear tensor and azimuthal velocity of fluids in the locally non - rotating frame ( lnrf ) by using keplerian angular velocity .
we derive two kinds of shear tensors in lnrf and blf ; in the first one the direction of fluid rotation is the same as that of black hole ( @xmath3 ) and in the second one the direction of fluid rotation is in opposite to that of black hole(@xmath4 ) .
we calculate the components of the shear tensor for two kinds of fluids in lnrf , blf and frf , these calculations show that in frf there are two nonzero components ( @xmath5 and @xmath2 components ) . but in previous papers ,
the only nonzero component in frf was @xmath2 component .
the @xmath5 component results from the relativistic calculations of the shear tensor which changes some components of the four velocity .
then , by using these shear tensor components we calculate the four velocity in lnrf , blf and density in all frames .
this paper s agenda is as follows .
we introduce the metric and reference frame in @xmath62 . in @xmath63
basic equations are given .
in @xmath64 the shear tensor is calculated in frf , lnrf and blf .
we derive four velocity in lnrf and blf in @xmath65 and the influence of two important parameters on density and four velocity can be seen in this section .
also , the influence of the @xmath5 component of shear tensor can be seen in this section .
summery and conclusion are given in @xmath66 .
for back ground geometry , we use the boyer - lindquist coordinates of the rotating black hole space time . in the boyer - lindquist coordinates ,
the kerr metric is : @xmath7 where @xmath8 and @xmath1 .
nonzero components of the lapse function @xmath9 , the shift vector @xmath10 and the spatial matrix @xmath11 are given in the geometric units as : @xmath12 here , we use geometric mass @xmath13 , @xmath14 , @xmath15 and @xmath16 .
the position of outer and inner horizons , @xmath17 , are calculated by inserting @xmath18 to get @xmath19 .
the angular velocity of the frame dragging due to the black hole rotation is @xmath20 , where m is the black hole mass , g is the gravitational constant and c is the speed of light and the angular momentum of the black hole , j , is described as : @xmath21 where @xmath22 .
similar to gammie & popham ( 1998 ) , we set @xmath23 for basic scalings .
the nonzero components of metric , @xmath24 , and its inverse , @xmath25 are calculated in appendix 1 . in our study , we use three reference frames .
the first one is the boyer - lindquist frame ( blf ) based on the boyer - lindquist coordinates describing the metric , in which , our calculations are done .
the second one is the locally non - rotating reference frame ( lnrf ) which is formed by observers with a future - directed unit vector orthogonal to @xmath26 . by using the boyer - lindquist coordinates ,
the lnrf observer is moving with the angular velocity of frame dragging ( @xmath27 ) .
the third frame is fluid rest frame ( frf ) , an orthonormal tetrad basis carried by observers moving along the fluid .
the physical quantities measured in lnrf are described by using the hat such as @xmath28 and @xmath29 and in frf by using parentheses such as @xmath30 and @xmath31 .
the transformation matrixes of frf , lnrf and blf are given in appendix 2 ( bardeen 1970 ; bardeen , press & teukolsky 1972 ; frolov & novikov 1998 ) .
the basic equations for the relativistic hydrodynamics are the baryon - mass conservation @xmath32 and the energy momentum conservation ; @xmath33 , where @xmath34 is the rest - mass density and @xmath35 is the energy - momentum tensor .
basic dynamical equations except the baryon mass conservation are calculated from the energy - momentum tensor , @xmath35 .
we use the energy - momentum tensor written as : @xmath36 where p is the pressure , @xmath37 is the relativistic enthalpy , u is the internal energy , @xmath38 is the viscous stress - energy tensor and @xmath39 is the heat - flux four vector .
the relativistic navier - stokes shear stress , @xmath38 , is written as ( misner , thorne & wheeler 1973 ) : @xmath40 where @xmath41 is the coefficient of dynamical viscosity , @xmath42 is the coefficient of bulk viscosity , @xmath43 is the projection tensor , @xmath44 is the expansion of the fluid world line , and @xmath45 is the shear tensor of the fluid which is calculated as : @xmath46 where @xmath47 is the four acceleration .
we study a stationary , axisymmetric and equatorially symmetric global accretion flow in the equatorial plane , i.e. , we assume @xmath48 .
we also assume that the effects of the bulk viscosity and heat - flux four vector is negligible . in the following sections
, we derive the basic equations using the vertical averaging procedures around the equatorial plane which were derived in , e.g. , gammie & popham ( 1998 ) .
the equation for the baryon mass conservation is written as : @xmath49 where @xmath50 is the four velocity and @xmath51 . by averaging the physical quantities around the equatorial plane and assuming constant @xmath52 ( the mass - accretion rate ) , we can write equation ( [ 7 ] ) as follows : @xmath53 where @xmath54 is half - thickness of the accretion disk in @xmath55 direction .
if we normalize the rest - mass density , @xmath34 , by setting @xmath56 , for calculating global structure of accretion flow , we have : @xmath57 also , by differentiating equation ( [ 11 ] ) we have : @xmath58 two specific killing vectors of kerr metric are @xmath59 and @xmath60 , which are used to derive the disk equations . at first
, angular momentum conservation can be derived by @xmath61 @xmath62 by vertically averaging equation ( [ 13 ] ) we have : @xmath63 where @xmath64 is the total inward flux of angular momentum .
this equation is similar to the angular momentum equation of gammie & popham ( 1998 ) .
another equation which can be obtained by using @xmath65 is : @xmath66 similarly , by vertically averaging and inserting constant @xmath67 , we have : @xmath68 this equation expresses the constancy of mass - energy flux @xmath67 in terms of the radius ; @xmath67 is the actual rate of change of the black hole mass .
if the fluid is cold and slow at large radiuses , then @xmath69 ( gammie & popham 1998 ) . the equation of local energy conservation is obtained from @xmath70 , as follows : @xmath71 where @xmath72 is defined as : @xmath73 and from equation ( [ 7 ] ) we have : @xmath74 therefore the energy equation can be written as : @xmath75 in this paper we do not derive the temperature , pressure and internal energy so that , we will not use this equation . if we want to derive these variables we must use a state equation and use the relation of shear tensor components .
lasota ( 1994 ) used the @xmath2 component of shear stress tensor as follows : @xmath76 @xmath77 where @xmath78 is the azimuthal component of velocity in the lnrf which is introduced by manmoto ( 2000 ) : @xmath79 where @xmath80 is defined as : @xmath81 abramowicz et al ( 1996 and 1997 ) , manmoto ( 2000 ) and others also used equation ( [ 21 ] ) for stress tensor . in 1994 ,
papaloizou & szuskiewicz introduced a non - relativistic causal viscosity which was used in relativistic form by peitz & apple ( 1997a , b ) , gammie & popham ( 1998)and popham & gammie ( 1998 ) in kerr metric and takahashi ( 2007b ) in kerr - schild metric .
they assumed that in the frf all components vanish , except @xmath82 .
the @xmath83 can be introduced in relativistic form as : @xmath84 where @xmath85 and @xmath86 is the equilibrium value of the stress tensor and @xmath87 is the relaxation time scale .
therefore , in steady state we have : @xmath88 we use a new method to derive the shear tensor approximately .
we calculate @xmath78 from equation ( [ 23 ] ) in lnrf assuming @xmath89 ( the azimuthal velocity is much greater than accretion velocity or radial velocity , except very close to the inner edge , and for fluids with a small viscosity ) . in order to simplify ,
we assume @xmath90 , where keplerian angular velocity @xmath91 , is defined as : @xmath92 therefore , for azimuthal velocity we have : @xmath93 first we calculate the @xmath94 . we can calculate four velocity in lnrf ( appendix 3 ) .
the four velocity for @xmath4 ( direction of fluid rotation is opposite to the black hole rotation ) is : @xmath95 calculations show that in lnrf @xmath96 and @xmath97 .
the shear tensor can be calculated by inserting the four velocity of lnrf in equation ( [ 6 ] ) .
the nonzero components of @xmath98 , @xmath99(@xmath100 ) and @xmath101 for @xmath4 in lnrf are : @xmath102 @xmath103 @xmath104 where @xmath105 . also for @xmath3 , following appendix 3 , the four velocity is : @xmath106 therefore , the nonzero components of @xmath107 and @xmath101 for @xmath3 are(similar to @xmath4 , @xmath108 ) : @xmath109 @xmath110 where @xmath111 . the shear tensor in blf can be derived from @xmath112 in which @xmath113 and @xmath114 are the transformation tensors(appendix 2 ) .
there are two nonzero components of shear tensor in blf , @xmath115 and @xmath116 , which are given in appendix 4 . also using transformation matrices , two nonzero components of shear tensor in frf
can be calculated including @xmath117 and @xmath118 which are given in appendix 4 as well .
assuming special values for @xmath41 ( for example in takahashi ( 2007b ) , @xmath119 and @xmath120 ) , we can solve equations ( [ 12 ] ) , ( [ 14 ] ) , ( [ 16 ] ) and ( [ 20 ] ) to derive four velocity , density , pressure and etc , numerically .
in numerical solutions we must insert values of physical variables on boundary conditions specially on horizon .
some of these boundary conditions are nonphysical , therefore we want to derive analytical solutions without any boundary conditions .
we use some non - relativistic relations to calculate four velocity and density in lnrf and blf analytically . at first
, the @xmath121 relation of stress tensor in takahashi ( 2007a ) is used in a relativistic disk , this relation is : @xmath122 similar to abramowicz et al.(1996 ) we assume @xmath123 , therefore from equation ( [ 5 ] ) by zero bulk viscosity and @xmath124 , we have : @xmath125 therefore , @xmath126 is @xmath127 for calculation the @xmath126 we use equations [ 02 ] and [ 01 ] with a simplification in @xmath128 and @xmath129 as : @xmath130 @xmath131 influences of this approximation in shear tensor components are seen in figure 1 for @xmath3 and @xmath132 .
solid curves have no simplifying and dotted curves are with this simplifying .
= 3.2 in = 2.3 in = 3.2 in = 2.3 in setting @xmath133 ( takahashi 2007a ) , the angular momentum can be calculated . from equations ( [ 34 ] ) and ( [ 14 ] ) , @xmath34 can be derived as : @xmath134 where @xmath135 is ( takahashi 2007a ) : @xmath136 which is the usual @xmath9 prescription for viscosity(shakura & sunyaev 1973 ) . in some papers @xmath54
is calculated by studying the vertical structure such as riffert & herold ( 1995 ) , abramowicz et al.(1997 ) and takahashi(2007b ) . in these papers @xmath54
was calculated by vertical averaging , then introduced by other variable such as @xmath137 , @xmath138 , @xmath139 , @xmath34 and etc .
if we use each of those relations of @xmath54 , we must solve all equations numerically .
but we want to derive an analytical solution therefore similar to takahashi(2007a ) we use a simple form of @xmath54 as : @xmath140 definitely , @xmath34 is the same in lnrf and blf and is shown in figure 2 for ( @xmath4 ) and ( @xmath3 ) by assuming @xmath141 .
= 3.1 in = 2.1 in = 3.1 in = 2.1 in also @xmath142 can be calculated from equation ( [ 11 ] ) and @xmath143 can be found from equation ( [ 16 ] ) .
assuming @xmath144 we have : @xmath145 we can calculate four velocity in blf using the transformation equations in appendix 2 , as : @xmath146 @xmath147 @xmath148 now , we may see ( @xmath149 ) , ( @xmath150 , @xmath151 ) and ( @xmath152)in figure 3 .
= 3.1 in = 2.1 in = 3.1 in = 2.1 in = 3.1 in = 2.1 in = 3.1 in = 2.1 in = 3.1 in = 2.1 in , = 3.1 in = 2.1 in four velocity and @xmath153 in blf(@xmath154 ) are shown in figure 4 setting @xmath141 .
= 3.1 in = 2.1 in = 3.2 in = 2.1 in = 3.2 in = 2.1 in = 3.2 in = 2.1 in = 3.2 in = 2.1 in = 3.2 in = 2.1 in = 3.2 in = 2.1 in = 3.2 in = 2.1 in in equation ( [ 35 * ] ) , @xmath155 is viscosity coefficient in @xmath9 prescription which influences on redistribution of energy and momentum .
the influence of these parameters are shown in figure 5 with @xmath141 = 3.2 in = 2.3 in = 3.2 in = 2.3 in the four velocity and density can be derived for various total inward flux of angular momentum ( for example @xmath156 and 3 ) which are shown in figure 6 .
= 3.2 in = 2.3 in = 3.2 in = 2.3 in = 3.2 in = 2.3 in = 3.2 in = 2.3 in = 3.2 in = 2.3 in in pervious papers such as abramowicz et al .
( 1996 and 1997 ) , gammie & popham ( 1998 ) , manmoto ( 2000 ) and takahashi ( 2007a , b ) , only the @xmath2 component of shear tensor was assumed to be nonzero in frf , but in section 4 we showed that the @xmath5 component is also nonzero .
the @xmath5 component of shear tensor in lnrf results from covariant derivative of @xmath143 . in general relativity the gravitating field causes time dilation . due to this time dilation ,
the coordinate time(t ) and proper time(@xmath157 ) are not the same(@xmath158 ) , therefor @xmath159 in lnrf can be derived as : @xmath160 @xmath161 is different in each radius and also tangent of @xmath143 creates the r - t component of shear tensor . to see the influence of this component ,
we derive four velocity with both @xmath5 component of shear tensor and without it .
if we set @xmath162 in the pervious section the figures are similar to that of the previous papers . from equations ( [ 35 ] ) and ( [ 35 ] ) it is clear that @xmath5 component affects @xmath138 in all frames because without this component , the second term of @xmath138 in equation ( [ 35 ] ) vanishes . if we calculate @xmath163 , the @xmath5 component also affects @xmath164 in blf .
the influences of @xmath5 component of shear tensor are shown in figure 7 , solid curves are with this component and dotted curves are without it .
= 3.2 in = 2.3 in = 3.2 in = 2.3 in = 3.2 in = 2.3 in = 3.2 in = 2.3 in = 3.2 in = 2.3 in = 3.2 in = 2.3 in obviously , @xmath5 component of shear tensor have more influence on @xmath138 and @xmath165 .
therefore , for deriving @xmath138 and @xmath165 we can not vanish the @xmath5 component .
in causal viscosity method , only @xmath2 component of shear tensor is assumed to be nonzero in frf .
we do not use this method for viscosity because our calculations show that there are two nonzero components of shear tensor in frf . in our method
, we use azimuthal velocity of lnrf and keplerian angular velocity , then we calculate all components of shear tensor in lnrf for two kinds of fluids ( rotation in the direction of black hole ( @xmath3 ) and rotation in the opposite direction of black hole(@xmath4 ) ) . using transformation tensor
we can calculate all components of shear tensor in blf and frf .
solid curves of figure 1 show unphysical treatments in shear tensor components close to the inner edge .
this unphysical treatment in @xmath166 may be concerned to assuming @xmath167 , therefore if we find a suitable @xmath168 this treatment may be resolved .
we may have a suitable @xmath168 if we put @xmath169 . according to equation ( [ 23 ] ) , the most important origin of unphysical treatments in @xmath170 is the @xmath171 term ( @xmath171 has a singularity in the inner edge ) .
therefor by using taylor expansion the unphysical treatment of @xmath166 and @xmath170 were solved as it can be seen in dotted curves of figure 1 .
we solved the hydrodynamical equations in lnrf with assuming @xmath123 , we derived the density and four velocity in blf and lnrf .
the density which is calculated analytically in lnrf is the same in all frames such as blf .
but comparison of four velocity in lnrf and blf ( figure 3 ) shows that @xmath172 and @xmath138 have a small difference in two frames . in equation ( [ 37 ] ) , @xmath173 is near 1 , specially in larger radiuses , therefore @xmath172 is similar in two frames .
following equation ( [ 36 ] ) , for @xmath138 we have two parts , the first part is @xmath174 and the second part is @xmath175 . because value of @xmath176 is greater than @xmath143(dotted curves in figure 3 ) , the first term has more influence on @xmath138 than the second term ( the second term is important just in inner radiuses ) .
@xmath176 and @xmath177 are different in two frames , because in equation ( [ 38 ] ) , @xmath178 . the previous equation is logical because the lnrf rotates with the angular velocity of the frame dragging ( the frame dragging velocity can not be seen in lnrf but it can be seen in blf ) , therefore @xmath177 is greater than @xmath176 in all radiuses . in section 5 , four
velocity was calculated analytically in lnrf and blf .
the sign of @xmath179 introduces two types of fluids , @xmath3 and @xmath4 , but it can be seen that for example , in figures 2 or 4 , in @xmath132 @xmath3 is similar to @xmath4 in @xmath180 ( and in @xmath181 @xmath3 is similar to @xmath4 in @xmath182 and so on ) therefore we use ( @xmath3 ) in figures 5 , 6 and 7 .
the effects of @xmath9 coefficient are shown in figure 5 .
it shows that if @xmath9 increases , then the shear viscosity grows , the radial four velocity will increase and then density decreases .
the @xmath5 component of the shear tensor in lnrf which results from the relativistic calculations(equation [ 6])is equal to covariant derivative of @xmath143 .
if we ignore the @xmath5 component of shear tensor ( the dotted curves in figure7 ) , the figures are similar to the figures of the previous works such as gammie & popham ( 1998 ) and takahashi ( 2007b ) .
but , with @xmath5 component of shear tensor ( the solid curves in figure 7 ) a few number of four velocities change in all frames , specially @xmath159 , @xmath165 , @xmath164 and @xmath183 will change more than the others .
the influences of the @xmath5 component on all physical variables can be derived if we solve all the equations of the disk with a state equation numerically .
when we have time dilation @xmath165 has to be greater than 1(@xmath184 ) . according to figure 4 , the effect of time dilation is much greater in the inner edge than the other places .
if we ignore the @xmath5 component of shear tensor or put @xmath185(dotted curve in the first panel of first column of figure 7 ) , it is equivalent to ignoring time dilation . equations ( [ 35 ] ) and ( [ 37 ] ) show that the @xmath5 component of shear tensor have no influence on @xmath142 and @xmath186 . near the black hole
@xmath165 increases , which causes a decreases in @xmath183 ( @xmath187 ) .
energy can be calculated from @xmath188 , where the first term of energy is related to @xmath138 .
figure 2 shows that @xmath34 is greater in inner edge than the other places and also is the same with the @xmath5 component of shear tensor and without it .
the @xmath138 which uses the @xmath5 component of shear tensor is greater than the @xmath138 without it .
therefore , the first term of energy with @xmath5 component of shear tensor is greater than the this term of energy without @xmath5 component . far from the black hole
the general relativistic influences are vanished , therefore in outer edge , the influences of @xmath5 component of shear tensor are too low ( as it is seen in outer radiuses of figure 7 ) .
if we want to calculate the temperature , pressure , internal energy , cooling and heating or radiation , we must use a suitable state equation with equations ( [ 11 ] ) , ( [ 14 ] ) , ( [ 16 ] ) and ( [ 20])(energy equation ) , then , to solve the equations numerically suitable boundary conditions are needed .
+ + * acknowledgments * we are grateful to the referee for a very careful reading of the manuscript and for his / her suggestions , which have helped us improve the presentation of our results .
abramowicz , m. a. , chen , x. , granath , m. , lasota , j .-
, 1996 , apj , 471 , 762 abramowicz , m. a. , lanza , a. , percival , m. j. , 1997 , apj , 479 , 179 bardeen , j. m. , 1970 , apj , 162 , 71 bardeen , j. m. , press , w. h. , teukolsky , s. a. , 1972 , apj , 178 , 347 chakrabarti , s. , 1996 , apj , 471 , 237 frolov , v. p. , novikov , i. d. , 1998 , black hole physics : basic concepts and new developments , kluwer academic gammie , c. f.,popham , r. , 1998 , apj , 498 , 313 lasota , j .-
1994 , in theory of accretion disks-2 , ed .
w. j. duschl , j. frank , f. meyer , e. meyer - hofmeister , w. m. tscharnuter manmoto , t. , 2000 , apj , 534 , 734 misner c. w. , thorne k. s. , wheeler j. a. , 1973 , gravitation .
freeman , san francisco papaloizou , j. c. b.,szuszkiewicz , e. , 1994 , mnras , 268 , 29 peitz , j. , appl , s. , 1997a , mnras , 286 , 681 peitz , j. , appl , s. , 1997b , mnras , 296 , 231 popham , r. , gammie , c. f. , 1998 , apj , 504 , 419 riffert , h. , herold , h. 1995 , apj , 450 , 508 shakura , n. i. , sunyaev , r. a. 1973 , a & a , 24 , 337 takahashi , r. , 2007a , a & a , 461 , 393 takahashi , r. , 2007b , mnras , 382 , 567
nonzero components of kerr metric are given as : @xmath189 and inverse components of metric @xmath25 are : @xmath190
components of @xmath191 connecting between blf ( boyer - lindquist frame ) and lnrf(locally non - rotating frame ) are calculated as : @xmath192 @xmath193 the transformation between lnrf and frf(fluid rest frame ) are as follows : @xmath194 @xmath195 where we use @xmath196 , then the lorentz factor @xmath197 is calculated as : @xmath198 in lnrf , three velocity components are calculated as : @xmath199 after some calculation we have ( @xmath200 ) : @xmath201
from equation ( [ 27 ] ) we have : @xmath202 first we derive four velocity for @xmath203 as @xmath204 in lnrf @xmath205 and in our study we put @xmath48 .
if we suppose @xmath206 ( as it was in pervious papers such as popham & gammie 1998 and takahashi 2007a , b ) , then @xmath207 and we have : @xmath208 also we know that @xmath209 therefore , @xmath210 for @xmath211 we have : @xmath212
nonzero components of shear tensor in blf can be calculated by @xmath112 where @xmath213 and @xmath214 are given in appendix b. therefore , these components for @xmath3 are : @xmath215 non - zero components of shear tensor in frf can be calculated by @xmath216 where @xmath217 and @xmath218 are transformation matrixes of appendix b. @xmath219 also , we have : | in this paper we solve the hydrodynamical equations of optically thin , steady state accretion disks around kerr black holes . here , fully general relativistic equations are used .
we use a new method to calculate the shear tensor in the lnrf ( locally non - rotating frame ) , blf ( boyer - lindquist frame ) and frf ( fluid rest frame ) .
we show that two components of shear tensor in the frf are nonzero ( in previous works only one nonzero component was assumed ) .
we can use these tensors in usual transonic solutions and usual causal viscosity , but we derive solutions analytically by some simplifications .
then we can calculate the four velocity and density in all frames such as the lnrf , blf and frf .
keywords : black hole accretion disks , relativistic disks , accretion disks , shear tensor , hydrodynamic . |
dynamics in strong fields has been an interesting subject in theoretical physics .
recently , this subject is being paid attention also in the experimental physics of creation of the quark gluon plasma . in high - energy heavy - ion collision experiments , at the so - called glasma stage @xcite just after the collision , longitudinal color electric and magnetic fields are expected to be produced in the context of the color glass condensate of order @xmath0@xmath1 gev in rhic and @xmath2 gev in lhc . in the peripheral collision , a strong magnetic field of order @xmath3 mev
would be induced .
the question is how the strong fields decay and the system is thermalized . in this work ,
we concentrate on how the strong fields decay into particles . for this purpose
, we first briefly review the schwinger mechanism in the coexistence of electric and magnetic fields .
we will point out that the vacuum immediately decays in the case of massless fermion and nonzero @xmath4 and @xmath5 . for simplicity
, we consider the case that the electric and magnetic fields are covariantly constant @xcite , i.e. , @xmath6=[d_\mu,\bm{b}]=\bm{0}$ ] , where @xmath7 is the covariant derivative with the gauge field @xmath8 .
the electric and magnetic fields are defined as @xmath9 and @xmath10 with @xmath11/g$ ] .
this is a generalization of constant fields in qed , @xmath12 , to the non - abelian fields .
for the covariantly constant fields , all the components of @xmath4 and @xmath5 can be diagonalized to be constant matrices in color space by a gauge transformation . without loss of generality
, one can also set @xmath13 and @xmath14 by choosing an appropriate lorentz frame and the coordinate axis .
the vacuum decay in an electric field was discussed by @xcite .
consider the vacuum persistency probability , which is defined by @xmath15 where @xmath16 and @xmath17 are infinite space volume and time length . @xmath18 and
@xmath19 are the in - vacuum and the out - vacuum defined at @xmath20 and @xmath21 , respectively .
if the vacuum is unstable , @xmath22 has a nonzero value , while , if the vacuum is stable , @xmath22 vanishes .
therefore , @xmath22 denotes magnitude of the vacuum decay per unit volume and time .
when @xmath22 is small , @xmath23 , so that @xmath22 is regarded as the pair production probability per unit volume and time . for qcd , the analytic formula of @xmath22 for the quark - pair creation in the covariantly constant
is given by @xcite @xmath24 where @xmath25 denotes the quark - mass matrix and the trace is taken over the indices of color and flavor .
this is a non - abelian extension of the following formula for qed @xcite : @xmath26 with the qed coupling constant @xmath27 .
note that the fermion pair creation formalism in the covariantly constant fields in qcd is similar to that in qed , so that we hereafter give the formula for qed , where we set @xmath28 and @xmath29 by a suitable axis choice and the parity transformation . in the absence of the magnetic field
, this formula reduces to the well - known result , @xmath30 if the masses are zero , @xmath22 has a _
finite _ value of @xmath31 .
the situation changes if the magnetic field exists . from eq .
( [ eq : w ] ) , @xmath22 diverges in the presence of the magnetic field . to see this , summing over all modes in eq .
( [ eq : w ] ) , we obtain for small @xmath25 as @xmath32 as @xmath33 , @xmath22 logarithmically diverges as @xmath34 next , let us consider the origin of the divergence of @xmath22 in terms of effective dimensional reduction in a strong magnetic field .
when a magnetic field exists , the spectrum of the transverse direction is discretized by landau quantization .
actually , the energy spectrum for @xmath35=0 is given by @xmath36 where @xmath37 correspond to the landau levels , and @xmath38 is the spin .
the system effectively becomes @xmath39 dimensional system with infinite tower of massive state : @xmath40 .
for the lowest landau level ( lll ) , @xmath41 and @xmath42 , the energy is @xmath43 .
this is the spectrum in @xmath39 dimensions .
this lll causes the divergence of @xmath22 as will be shown below .
the divergence of @xmath22 does not mean the divergence of the infinite pair production per unit space - time .
the divergence of @xmath22 rather implies that the vacuum always decays and produces pairs of fermion .
the question is where the vacuum goes . in the coexistence of @xmath44 and @xmath35
, one can obtain the probability of the @xmath45 pairs of fermion with lll as @xmath46 .
\end{split } \label{eq : numberofpairs}\ ] ] the vacuum persistency probability corresponds to all @xmath47 s being zero in eq .
( [ eq : numberofpairs ] ) , and @xmath22 is equal to eq .
( [ eq : wlll ] ) , so that @xmath22 diverges at @xmath48 . at @xmath48 ,
this probability is finite only if the following equation is satisfied : @xmath49 therefore , the number of the particle with the lll is restricted by eq .
( [ eq : finiteprobabilitycondition ] ) , and linearly increases with time .
the higher landau levels give heavy effective masses of order @xmath50 , so that all the contributions to the pair productions from such modes are suppressed .
the total number of the particle pairs can be calculated : @xmath51 at @xmath52 , @xmath53 .
the contribution of lll is obtained as @xmath54 which is equal to taking @xmath55 in eq .
( [ eq : numberofpairseb ] ) . .
the solid line denotes the contribution from the lll , and the dotted line denotes the contribution from all modes . ] in fig .
[ fig : totalnumber ] , the total number of the particle for the full contribution and lll contribution are shown .
the lll dominates for @xmath56 , so that the effective model for the lll works well for @xmath56 .
in this section , we study particle productions coming from the lll for qed taken into account the back reaction . for this purpose , we consider lll projected theory , that is , the wave function of the fermion is projected to the lll state .
the wave function of the lll is @xmath57 where @xmath58 denotes the angular momentum in @xmath59 direction for the lll , and the energy is degenerate for @xmath58 .
one can decompose the fermion field into the longitudinal mode and the transverse mode in a suitable representation as @xmath60 where @xmath61 is the two component dirac field in 1 + 1 dimensions .
then the fermion action of qed in @xmath62 dimensions reduces to that of non - abelian gauge theory in @xmath39 dimensions : @xmath63 where @xmath64 and @xmath65 are the gamma matrices in @xmath39 dimensions and @xmath66 .
the covariant derivative is defined by @xmath67 with @xmath68 @xmath69 corresponds to the gauge field in @xmath70 gauge theory , since @xmath69 is an hermite matrix , @xmath71 , and the indices @xmath58 and @xmath72 run from @xmath73 to @xmath74 .
to simplify the situation , we assume that the @xmath75 and @xmath76 do not depend on the transverse directions , @xmath77 and @xmath78 .
then the @xmath58 dependence can be factorized out : @xmath79 and @xmath80 .
the action in eq .
( [ eq : action ] ) becomes @xmath81 where @xmath82 is the volume of the transverse directions .
this action is nothing but that in 1 + 1 qed , i.e. , the schwinger model , except for the overall factor @xmath83 .
the exact solution of the effective action for the fermion is known as plotted against time . both values are normalized by that at maximum values.,title="fig : " ] plotted against time .
both values are normalized by that at maximum values.,title="fig : " ] @xmath84 where @xmath85 denotes for longitudinal directions , @xmath86 and @xmath59 .
@xmath87 denotes the effective photon mass , @xmath88 .
equation ( [ eq : effectiveaction ] ) is manifestly gauge invariant . in the lorenz gauge
, it reduces to the result by @xcite .
the mass @xmath87 is induced by the axial anomaly , of which effect is called `` dynamical higgs effect . ''
this mass generation is related to the fact @xmath89 as @xmath33 . using this form
, we can calculate the fermion and axial currents , @xmath90 where we choose the lorenz gauge , @xmath91 .
the divergence of the axial current leads the axial anomaly in @xmath39 dimensions except for the overall factor @xmath92 : @xmath93 this relation is nothing but axial anomaly in @xmath62 dimensions . since the effective action in eq .
( [ eq : effectiveaction ] ) has a quadratic form in @xmath94 , the equation of motion for the photon can be solved .
for example , the electric field of @xmath59 direction is @xmath95 where @xmath96 .
the currents satisfy @xmath97 and @xmath98 .
we show the electric field and the number density for the spatially homogeneous case , @xmath99 , as a function of time in fig .
[ fig : dynamics ] .
the electric field oscillates with a frequency @xmath100 . in this case ,
@xmath101 , but @xmath102 .
the number density of pairs is equal to @xmath103 .
these results agree with the previous works @xcite .
the generalization to non - abelian theories is straight forward if the magnetic field is enough strong .
the fermion determinant becomes wess - zumino - witten action .
in this work , we have discussed the vacuum decay in strong electric and magnetic fields . when the fermion is massless , the vacuum persistency probability per unit time and volume becomes zero , and hence @xmath22 diverges .
the origin of the divergence is from discretized spectra of transverse directions and the lowest landau level .
the lll level dominates for @xmath56 . with the lll projection ,
we have analytically calculated the effective action in this situation , and reproduced the previous numerical results .
since the effective theory of the lll is solvable , there is no chaotic behavior nor thermalization .
the thermalization does not happen in the lll - dominant process . in the case of qcd ,
the gluon is more interesting because of its self - interaction .
the landau quantization causes instability because the helicity of gluon is one .
inserting @xmath104 in eq .
( [ eq : energy ] ) , the energy becomes imaginary when @xmath105 , of which instability is known as nielsen - olesen instability .
the situation is similar to quenching phenomenon in general phase transition , where a temperature suddenly changes .
in such a situation , a phase separation occurs .
the same phenomena would happen in relativistic heavy - ion collisions , because the electric and magnetic fields are suddenly induced by the collision , and the perturbative vacuum of gluon is unstable .
although dynamics of unstable gluon vacuum is very interesting , we leave this topic in the future work . 9 t. lappi and l. mclerran , nucl .
phys . a 772 , 200 ( 2006 ) .
h. suganuma , t. tatsumi , prog .
theor . phys .
90 , 379 ( 1993 ) .
w. heisenberg and h. euler , z. phys .
98 , 714 ( 1936 ) .
j. schwinger , phys .
82 , 664 ( 1951 ) .
n. tanji , ann .
324 , 1691 ( 2009 ) ; ibid 325 , 2018 ( 2010 ) .
j. schwinger phys .
128 , 2425 ( 1962 ) ; in _ theoretical physics _ , trieste lectures 1962 ( iaea , vienna , 1963 ) p.89 .
a. iwazaki , phys .
c 80 , 7 ( 2009 ) . | we discuss fermion pair productions in strong electric and magnetic fields .
we point out that , in the case of massless fermions , the vacuum persistency probability per unit time and volume is zero in the strong electric and magnetic fields , while it is finite when the magnetic field is absent .
the contribution from the lowest landau level ( lll ) dominates this phenomenon .
we also discuss dynamics of the vacuum decay , using an effective theory of the lll projection , taking into account the back reaction . |
the sum - over - states ( sos ) expressions have been used for more than three decades in the study of nonlinear optical phenomena , and are perhaps the most universally used equations in molecular nonlinear optics .
the sum - over - states expression is obtained from quantum perturbation theory and is usually expressed in terms of the matrix elements of the dipole operator , @xmath0 , and the zero - field energy eigenvalues , @xmath1.@xcite + the sos expressions for the first and second hyperpolarizability derived by orr and ward using the method of averages@xcite are often used because they explicitly eliminate the unphysical secular terms that are present in other derivations.@xcite these secular - free expressions contain summations over _ all _ excited states .
+ finite - state approximations are used to apply the theory to experimental results .
oudar and chemla studied the first hyperpolarizability of nitroanilines by considering only two states , the ground and the dominant excited state.@xcite although the general validity of this `` two - level '' model has been questioned , especially in its use for extrapolating measurement results to zero frequency , the approximation is still widely used in experimental studies of the nonlinear properties of organic molecules .
+ several approaches have been used to develop approximate expressions for the second - hyperpolarizability in the off - resonance regime.@xcite while such approximations are helpful , they systematically ignore some of the contributions to the sos expression . as our goal is to derive a general expression that is equivalent to the traditional sos one
, we choose not to make any assumptions a priori about what type of contributions dominate the response .
furthermore , including all the possible contribution is necessary to properly describe the on - resonance behavior , even when only few states contribute to the response.@xcite + in 2005 , kuzyk used the generalized thomas - kuhn sum rules to relate the matrix elements and energies involved in the general orr and ward sos expression for the first hyperpolarizability , and introduced a new and compact sos expression that does not depend explicitly on dipolar terms.@xcite since the thomas - kuhn sum rules are a direct and exact consequence of the schrdinger equation when the hamiltonian can be expressed as @xmath2 , it follows that the new sos expression is as general as the original , converges to the same results , and by virtue of its compactness may be more appropriate for the analysis of certain nonlinear optical properties.@xcite indeed , champagne and kirtman used a comparison between the dipole - free and standard sos expressions to study the convergence of molecular - orbital calculations.@xcite in this work , we use the same principle to derive a compact and general dipole - free expression for the second hyperpolarizability .
while our method can be applied to non - diagonal components of the second hyperpolarizability , for simplicity we will focus on the diagonal component . the sos expression for the diagonal term of the second hyperpolarizability @xmath3 as derived by orr and ward in 1971
is given by:@xcite @xmath4 where @xmath5 is the magnitude of the electron charge , @xmath6 the @xmath7 matrix element of the position operator and @xmath8 ( @xmath9 ) are the frequencies of the photons with @xmath10 . the bar operator is defined as : @xmath11 the dispersion of @xmath3 is given by @xmath12 and @xmath13 which are defined as follows : @xmath14 @xmath15 where spontaneous decay is introduced by defining complex energies : @xmath16 where @xmath17 is the energy different between the @xmath18 excited state and the ground state , and @xmath19 is the inverse radiative lifetime of the @xmath18 state . to obtain a dipole - free expression for the second hyperpolarizability we begin by separating explicitly dipolar terms from dipole - free terms in the first term of eq .
[ dipolefree : eq : gammaxxxx ] , @xmath20 the second term in eq .
[ dipolefree : eq : gammaxxxx ] is already dipole - free .
+ it should be noted that for non - dipolar systems ( such as octupolar chromophores ) , with @xmath21 , only the last term in eq .
[ eq : gsplit ] contributes to the second hyperpolarizability .
the generalized thomas - kuhn sum rules can be used to obtain a relationship between the explicitly dipolar terms in terms of only non - dipolar terms:@xcite @xmath22 we stress that the only assumption made in the derivation of eq .
[ eq : diagonal ] is that the sum rules hold , which is the case when the unperturbed hamiltonian describing the system is conservative .
+ substituting eq .
[ eq : diagonal ] into eqs .
[ eq : gsplit ] and [ dipolefree : eq : gammaxxxx ] yields the dipole - free expression for the second hyperpolarizability : @xmath23
so , equation [ dipolefree : eq : final2 ] is as general as the traditional sum - over - states expression.@xcite
it is useful to compare the convergence between the dipole - free expression for the second hyperpolarizability ( eq . [ dipolefree : eq : final2 ] ) with the traditional orr and ward sos expression ( eq .
[ dipolefree : eq : gammaxxxx ] ) for various systems . in this section
we will compare these expressions as a function of wavelength for two model systems .
mathematically , both expressions are equivalent , as long as all excited states of the system are included in the sum , so this exercise will determine how many states are required for convergence . since in practice , the sum - over - states expressions must be truncated , it is critical to understand the effect of discarding terms on the nonlinear susceptibility .
we also apply this new expression to calculate the fundamental limits of @xmath3 , and show that the results agree with those obtained using the standard sos expression .
+ we begin by first calculating the fundamental limit of @xmath3 starting from the dipole - free expression .
the analogous calculation has already been performed using the traditional orr and ward sos expression,@xcite so we can check whether or not the two results are the same
. a different set of results would suggest that the method used in calculating the fundamental limits does not hold .
+ according to the three - level ansatz,@xcite when near the fundamental limit , only three - levels contribute to the nonlinear response , eq . [ dipolefree : eq : final2 ] becomes : @xmath24 off - resonance , the dispersion terms ( eqs .
[ eq : dispersion3 ] and [ eq : dispersion4 ] ) simplify to : @xmath25 and @xmath26 and the relationships between the first transition dipole moments can be evaluated from the thomas - kuhn sum rules:@xcite @xmath27 \label{x_02}\\ |x_{12}|^{2 } \leq \frac{e_{10}}{(e_{20}-e_{10 } ) } \left [ |x_{01}^{max}|^{2}+|x_{01}|^{2 } \right ] , \label{x_12}\end{aligned}\ ] ] with@xmath28 where @xmath29 is the number of electrons in the system .
+ introducing the dimensionless quantities : @xmath30 the off - resonance diagonal component of the second hyperpolarizability can be written as : @xmath31 where @xmath32 is defined by : @xmath33 the second hyperpolarizability scales as @xmath34 , the square of the number of delocalized electrons , and as @xmath35 , the fifth power of the wavelength of maximum absorption . while in a three - level model ,
the expression for the first hyperpolarizability , which is analogous to equation [ eq : gamma3fx ] , explicitly separates into a product of a function of the transition dipole moment @xmath36 and excited state energies ( i.e. @xmath37),@xcite this is not possible for the case of the second hyperpolarizability . to optimize the second hyperpolarizability the function @xmath38 has to be optimized as function of the two parameters @xmath39 and @xmath40 .
the behavior of the function @xmath41 ( as given by eq .
[ eq : gamma3fx ] ) as a function of the parameters @xmath40 and @xmath39 is shown in fig .
[ fig : gamma3fx ] .
the function is maximized when both @xmath42 ( i.e. the second excited state energy level is far away from the first excited energy level ) and @xmath43 ( i.e. the oscillator strength is concentrated in the second transition dipole moment , @xmath44 ) .
when the function is optimized we obtain the quantum limit : @xmath45 @xmath32 ( as given by eq .
[ eq : gamma3fx ] ) spanning the full allowable range of the parameters @xmath40 and @xmath39 ( defined in eqs .
[ e_ratio ] and [ x_dim ] ) .
the function is optimized when simultaneously @xmath46 and @xmath43 . as @xmath47 and @xmath46 or @xmath48 the function becomes negative , corresponding to a negative value of the second hyperpolarizability . ]
this result agrees with the quantum limit obtained from the traditional orr and ward sos expression.@xcite + thus , we can conclude that when only three levels contribute to the nonlinear response , the dipole - free and the traditional sos expressions for the second hyperpolarizability - when simplified using the sum rules - become the same , leading to the same quantum limits .
we should point out that the quantum limits are obtained by assuming that the response is dominated by the contributions of three overlapping states , an ansatz that has been extensively verified numerically using monte carlo methods@xcite as well as potential energy optimization.@xcite there are no assumptions about the symmetry properties of the states .
however , this does not imply that symmetry plays no role in he optimization of the second hyperpolarizability .
the symmetry properties of the system will determine whether or not the optimal distribution of excited energies and transition dipole moments can be achieved .
mathematically , symmetries will impose further constraints on the parameters , which will make @xmath3 smaller .
we note that the quantum limit is negative for the centrosymmetric system , and is one - quarter of the positive limit that is obtained for an asymmetric molecule .
+ in this section , we test the convergence of the expressions in the case of two exactly solvable quantum mechanical systems : the `` particle in a box '' and the `` clipped harmonic oscillator '' . for simplicity
, we will first perform our calculations in the off - resonance regime .
+ the unperturbed states that we will use for our calculation of the second hyperpolarizability are the solutions of the one - dimensional time - independent schrdinger equation : @xmath49 + the potential that characterizes the particle in a box is zero inside the box of length @xmath50 and infinite , otherwise .
the solutions are given by : @xmath51 with @xmath52 the corresponding energies are : @xmath53 where @xmath54 is the mass of the particle ( in this case the electron mass ) .
these solutions are substituted into the expressions for the diagonal component of the second hyperpolarizability ( eqs .
[ dipolefree : eq : gammaxxxx ] and [ dipolefree : eq : final2 ] ) to study the convergence of both series as a function of number of excited levels included in the sum .
+ in the off - resonance regime , eq . [ dipolefree : eq : gammaxxxx ] becomes : @xmath55 and eq .
[ dipolefree : eq : final2 ] is given by : @xmath56 the numerical evaluation of the diagonal component of the second hyperpolarizability as given by eq .
[ eq : sosoff ] or eq .
[ eq : dipoleoff ] respectively , is performed by dividing every contribution in the sum by the quantum limit ( eq . [ eq : theglimit ] ) . for the particle in a box ,
a general term in the sum can be rewritten as : @xmath57 where we have used defined the following dimensionless functions : @xmath58 the convergence of the two series in the off - resonance regime is shown in fig .
[ fig : particle ] . when including the contribution of the first 10 states , the relative difference between traditional sos expression for the second hyperpolarizability ( eq . [ dipolefree : eq : gammaxxxx ] ) and the dipole - free expression ( eq .
[ dipolefree : eq : final2 ] ) is of the order of @xmath59 . with 50 states ,
the two expressions converge to the same value of : @xmath60 interestingly , the average value ( also shown in fig .
[ fig : particle ] ) is more accurate when as few as three states are included in the sum .
this is because , for the particular case of the particle in a box , if not enough states are included in the sums , the traditional sos expression tends to underestimate while the dipole - free expressions tends to overestimate @xmath3 . in this case , with few excited states , using the average value yields a more reliable estimation of the second hyperpolarizability .
this same result was found for the first hyperpolarizability@xcite and for studies used in modeling real molecules.@xcite convergence of the normalized sos expressions as a function of the number of states in the sum for the particle in a box . both the traditional sos expression ( eq . [ dipolefree : eq : gammaxxxx ] ) and the dipole - free expression ( eq . [ dipolefree : eq : final2 ] )
converge quickly to the exact value given by eq .
[ finalvalue ] . for completeness
, we also plot the normalized average value of the two expressions , which converges more rapidly as a function of the number of states than the traditional and dipole - free expressions alone .
the inset shows a mgnified view of the region indicated by the dash boxes . ]
another exactly solvable system is the clipped harmonic oscillator , whose potential energy function is given by:@xcite @xmath61 where @xmath62 has dimensions of frequency .
+ introducing the dimensionless variable : @xmath63 the solutions are expressed as : @xmath64 where @xmath65 is the @xmath66 order hermite polynomial , and the energies are given by : @xmath67 with @xmath52 + for the clipped harmonic oscillator , a general term in the sum can be rewritten as : @xmath68 where we have defined the following dimensionless function : @xmath69 the convergence of the two series in the off - resonance regime is shown in fig . [
fig : cho ] . in this case
, it takes more terms to reach convergence than for the particle in a box . for 50 states , the relative difference between the traditional sos expression ( eq .
[ dipolefree : eq : gammaxxxx ] ) and the dipole - free expression ( eq . [ dipolefree : eq : final2 ] ) is of the order of @xmath70 .
the traditional sos expression converges faster to the final value : @xmath71 convergence of the normalized sos expressions as a function of the number of states in the sum for the clipped harmonic oscillator .
both the traditional sos expression ( eq . [ dipolefree : eq : gammaxxxx ] ) and the dipole - free expression ( eq . [ dipolefree : eq : final2 ] ) converge slowly to the exact value given by eq .
[ finalcho ] . for completeness
, we also plot the normalized average value of the two expressions .
the inset shows a magnified view of the region indicated by the dash boxes . ] in this section , we investigate the convergence of the dispersion of the two forms of the second hyperpolarizability . in particular ,
we will treat two separate cases : two photon absorption ( tpa ) - which is related to the imaginary part of @xmath3 and used in a broad range of applications such as photodynamic cancer therapies and 3d photolithography,@xcite and the optical kerr effect(oke ) - which is related to the real part of @xmath3 and widely used in characterizing materials with potential applications in all - optical switching.@xcite we begin by evaluating eqs .
[ eq : dispersion3 ] and [ eq : dispersion4 ] and including only the first 6 excited states .
+ to get the typical qualitative behavior of real large chromophores , we choose @xmath72 .
all the linewidths are given by @xmath73 which is also typical for organic chromophores .
we normalize the second hyperpolarizability by dividing by the off - resonance limit for @xmath72 .
[ fig : distpapb ] shows the dispersion predicted by the traditional sum - over - states expression and the dipole - free expression for the imaginary part of the second hyperpolarizability as s function of the incident photon energy .
note that @xmath74 $ ] is related to the two photon absorption cross - section .
the agreement between the two expressions is excellent everywhere with the exception of the third resonance ( see inset in fig . [
fig : distpapb ] ) , where the two differ by less than 20% .
+ the normalized imaginary part of the two photon absorption second hyperpolarizability as a function of the incident photon energy for a 7-level model for the particle in a box using the traditional sum - over - states and the dipole - free expressions .
the inset shows a magnified view of the region indicated by the dashed box . @xmath75 . ]
next we compare the two expressions for the real part of the second hyperpolarizability - which is related to the optical kerr effect - as a function of the fundamental photon energy for the particle in a box .
the results are plotted in fig .
[ fig : diskerrpb ] . in this case , for 6 excited states , the two expressions differ by as much as a factor of 2 in some regions ( see insets ) .
although the qualitative behavior is the same , this type of discrepancy should be considered when experimental data is analyzed using a limited number of terms in the sum - over - states expressions .
+ the normalized real part of the kerr effect second hyperpolarizability as a function of the incident photon energy for a 7-level model for the particle in a box using the traditional sum over states and the dipole - free expressions .
the insets show a magnified view of the region indicated by the dashed boxes . @xmath75 . ]
the normalized imaginary part of the two photon absorption second hyperpolarizability as a function of the incident photon energy for a 7-level model for the clipped harmonic oscillator using the traditional sum - over - states and the dipole - free expressions .
the inset shows a magnified view of the region indicated by the dashed boxes . @xmath75 . ]
next we consider the dispersion predicted by the traditional sum - over - states expression and the dipole - free expression for the imaginary part of the two photon absorption second hyperpolarizability as a function of the fundamental photon energy for the clipped harmonic oscillator .
the results are plotted in fig .
[ fig : distpacho ] .
the agreement between the two expressions is good , differing only near the resonances .
finally , fig .
[ fig : dispkerrcho ] shows the dispersion predicted by the two expressions for the kerr effect second hyperpolarizability as a function of the fundamental photon energy for the clipped harmonic oscillator .
for 6 excited states , the two expressions differ substantially only in the vicinity of a resonance ( see insets ) .
the normalized real part of the kerr effect second hyperpolarizability as a function of the incident photon energy for a 7-level model for the clipped harmonic oscillator using the traditional sum - over - states and the dipole - free expressions .
the insets show a magnified view of the region indicated by the dashed boxes . @xmath75 . ]
in this section we will study the convergence of the dipole - free series expression ( eq .
[ dipolefree : eq : final2 ] ) as a function of the number of excited states included in the sum .
the convergence is studied as a function of photon energy for two photon absorption and the kerr effect , using the particle in a box and the clipped harmonic oscillator as model quantum systems .
again , in order to get the typical qualitative behavior of large real chromophores , we choose @xmath72 , and all the linewidths are given by @xmath73 which is also typical for organic chromophores . also , we normalize the second hyperpolarizability by dividing by the off - resonance limit for @xmath72 .
+ in all the cases it is found that after including about 15 excited states in the sum the expression converges even in the vicinity of a resonance . in order to get a better understanding of the convergence behavior of the expressions we will use the particle in a box model as a test for convergence , once we take into account symmetry considerations . + for clarity , we will look again to the expression for the second hyperpolarizability that separates explicitly dipolar terms from dipole - free terms : @xmath76 for a system whose symmetry demands that all dipole moments vanish , i.e. @xmath21 for @xmath77 , such as the the centrosymmetric system of a particle in a box or molecules with octupolar symmetry ( i.e. no dipole moment but non - centrosymmetric ) , the first three terms must each vanish .
numerically , when we use the dipole - free expression to model such systems that are centrosymmetric , which demands that all dipole moments vanish , the contributions from the first three terms are precisely zero only when an infinite number of terms are included in the sum .
+ it is useful to consider how many states are needed in order for each of these dipolar terms , written in our new non - dipolar form , to vanish .
we define the partial sums @xmath78 , @xmath79 and @xmath80 and @xmath81 as : @xmath82 clearly , for systems where there is no change in the dipole moment between the ground and excited states , the dipole - free expression will converge when simultaneously : @xmath83 , @xmath84 and @xmath85 and @xmath86 .
we will test these conditions numerically using the particle in a box as a model of a centrosymmetric potential with no dipole moment .
+ first we consider two photon absorption and study the convergence of @xmath78 ( eq . [ eq : s1 ] ) , @xmath79 ( eq . [ eq : s2 ] ) , @xmath80 ( eq . [ eq : s3 ] ) and @xmath81 ( eq . [ eq : s4 ] ) as a function of the number of excited states for a range of different photon energies .
the results are shown in fig .
[ fig : tpapbcombined ] .
clearly , for all cases , few excited states ( from 2 to 5 ) are needed to get good convergence when away from resonance . while @xmath78 also converges in the resonant regime after 10 excited states are included more excited states ( up to 20 )
are required for convergence of @xmath79 and @xmath80 near resonance ( see insets ) .
finally we look at the convergence of @xmath81 ( eq . [ eq : s4 ] ) , which as we have seen , converges to the exact value of the second hyperpolarizability for systems with no permanent dipole moment , such as the particle in a box .
surprisingly , this expression is shown to converge rapidly as a function of number of excited states included in the sum , even on resonance .
in fact , it is clear from the plot that only 2 excited levels might suffice to study the qualitative behavior of the second hyperpolarizability even close to the resonances
. + similarly , we next consider the kerr effect and study the convergence of @xmath78 ( eq . [ eq : s1 ] ) , @xmath79 ( eq . [ eq : s2 ] ) and @xmath80 ( eq . [ eq : s3 ] ) and @xmath81 ( eq . [ eq : s4 ] ) as a function of the number of excited states for a range of different photon energies . the results are shown in fig .
[ fig : kerrpbcombined ] .
again , few excited levels are needed for convergence away from resonance , although more excited states ( up to 20 ) are required for convergence near resonance . as in the case of the two photon absorption second hyperpolarizability , @xmath81 ( eq . [ eq : s4 ] ) , which is the only terms that contributes to the second hyperpolarizability for the centrosymmetric potential given by the particle in a box
is shown to converge rapidly as a function of number of excited states included in the sum , even in the resonant regime .
+ the normalized real part of the partial sums @xmath78 ( top - left ) , @xmath79 ( top - right ) , @xmath80 ( bottom - left ) and @xmath81 ( bottom - right ) for the kerr effect second hyperpolarizability as a function of the incident photon energy .
the number of states included in the sum is shown in the figure legends .
the particle in a box is used as a model centrosymmetric system with vanishing dipole moments , and therefore as the sums converge we must have @xmath83 , @xmath87 , @xmath88 and @xmath89 .
the insets show a magnified view of the region indicated by the dashed boxes . ] for completeness , we also study the convergence of the expressions for the two photon absorption process and the kerr effect using the clipped harmonic oscillator as a quantum model . as shown by the results in fig .
[ fig : chocombined ] the expressions do not converge as rapidly as did the particle in a box ; but when 15 excited states are included in the sum , the spectral features do not change quantitatively .
+ the normalized imaginary part of the two photon absorption second hyperpolarizability ( top ) and the normalized real part of the kerr effect second hyperpolarizability(bottom ) as a function of the incident photon energy .
the number of states included in the sum is shown in the figure captions .
the clipped harmonic oscillator is used as a model , with @xmath75 .
the insets show a magnified view of the regions indicated by the dashed boxes . ]
all of our results show that the dipole - free expression converges even in the resonant regime when enough excited states ( up to 20 ) are included in the expression .
we have also shown that for systems with no dipole moments ( such as octupolar molecules or the particle in a box ) the dipole - free expression for the second hyperpolarizability collapses to eq .
[ eq : s4 ] ( which we will call the _ reduced _ dipole - free expression ) , since all the other terms vanish . in this case , the reduced dipole - free expression converges when only a few excited states are included in the sum .
we have developed an expression that eliminates the explicit dependence on dipolar terms but is physically equivalent to the traditional sos expression for the second hyperpolarizability .
the equivalence between the dipole - free and the traditional sos expressions is demonstrated by calculating the quantum limits and studying the convergence of the series with the exact wavefunctions of two quantum systems : the particle in a box and the clipped harmonic oscillator . in both cases , when a large number of states is included , the two expressions are identical .
however , the average of the two expressions converges faster than the individual expressions .
+ since the average between the two expressions appears to be a better approximation to molecular dispersion , the average may make it possible to use limited - state models when interpreting experimental dispersion data . since accurate measurements of transition moments between excited states are difficult and tedious , the averaged second hyperpolarizability can be a useful tool for modeling the second hyperpolarizability when only limited information is available about the excited states of a particular system .
+ to test the convergence between the two expressions , we have evaluated them in the resonant regime in two model systems : the particle in a box - which is a symmetric potential with no change in dipole moment ; and the clipped harmonic oscillator - an asymmetric potential .
this allows us to determine the role of symmetry . in both cases ,
we study the dispersion of the second hyperpolarizability near resonance , where - based on the different energy denominators - one would expect the differences between the two expressions to be the least consistent .
the reduced dipole - free expression has been introduced for systems with no dipole moment .
such an expression might be most appropriate when experimental results are interpreted since it requires the minimum number of molecular parameters .
+ in conclusion , the dipole - free expression is an alternative to the traditional sos expression that increases the theoretical pallet available to quantum chemists .
it is more direct in certain theoretical problems such as its application to the derivation of a more - rigorous calculation of the fundamental limits of the third - order susceptibility .
it provides a tool to assess the convergence of truncated sos calculations , can be used to determine the accuracy of molecular - orbital calculations of nonlinear susceptibilities , and can be used to refine limited - state models to interpret experimental results . and , it may be more naturally applicable to the analysis of specific systems such as octupolar structures.@xcite + * acknowledgements : * jpm acknowledges the fund for scientific research flanders ( fwo ) and the support from the division of molecular and nanomaterials at the department of chemistry in kuleuven .
mgk thanks the national science foundation ( ecs-0354736 ) and wright paterson air force base for generously supporting this work . | the generalized thomas - kuhn sum rules are used to eliminate the explicit dependence on dipolar terms in the traditional sum - over - states ( sos ) expression for the second hyperpolarizability to derive a new , yet equivalent , sos expression .
this new dipole - free expression may be better suited to study the second hyperpolarizability of non - dipolar systems such as quadrupolar , octupolar , and dodecapolar structures .
the two expressions lead to the same fundamental limits of the off - resonance second hyperpolarizability ; and when applied to a particle in a box and a clipped harmonic oscillator , have the same frequency - dependence .
we propose that the new dipole - free equation , when used in conjunction with the standard sos expression , can be used to develop a three - state model of the dispersion of the third - order susceptibility that can be applied to molecules in cases where normally many more states would have been required .
furthermore , a comparison between the two expressions can be used as a convergence test of molecular orbital calculations when applied to the second hyperpolarizability . |
methanol masers in the @xmath0 line at 84521.21 mhz were found by batrla and menten ( 1988 ) and menten ( 1991 ) towards ngc 2264 , omc-2 , and dr 21 , but no extended survey in this line had been done .
the @xmath0 transition belongs to the class i ( menten , 1991 ) .
its excitation is similar to that of the @xmath2 and @xmath3 transitions .
since methanol masers emit in several lines of the same class , we expect the detection of a fairly large number of maser sources at 84.5 ghz .
their parameters should be taken into account when modeling maser sources .
therefore , we made a survey of known class i maser sources at 84.5 ghz .
the observations were carried out in may 1997 and march 2000 with the millimetre - wave telescope of the onsala space observatory . a sample of 13 sources at 84.5 ghz was observed in june 2000 with the 12-m nrao telescope at kitt - peak in remote mode from astro space center .
emission was detected in 51 of the 54 sources observed .
the spectra are markedly different from those of the strongest class i transition , @xmath4 at 44.1 ghz . at 44.1 ghz ,
most of the sources from our sample have bright and narrow maser features , whereas broad quasi - thermal components dominate at 84.5 ghz , and narrow ( @xmath5 km / s ) features are present in the spectra of only 17 of the 51 detected sources ( fig .
1 ) . however , it is possible that at least some of the quasi - thermal lines contain narrow maser components .
the shape of the 84.5 ghz spectra closely resembles the shape of the spectra of the same sources in the @xmath1 ( valtts et al .
1995 ) and @xmath3 ( slysh et al .
1999 ) transitions at 95.2 and 132.8 ghz , respectively .
the relationships between the integrated intensities of thermal lines at 84.5 , 95.2 and 132.8 ghz can be fitted by the equations @xmath6 and @xmath7 here @xmath8 is the main - beam brightness temperature .
the relative decrease of the line intensities at 132.8 , and especially at 95.2 ghz , is probably connected with the decrease of level population with increase of their energies : at a gas temperature of 35 k the population of the @xmath9 level is about 40% of the population of the @xmath10 level , making it possible to explain the relationships obtained .
note the detection of narrow features at 84.5 and 95.2 ghz towards the young bipolar outflow l 1157 . unlike other methanol masers , which are associated with high - luminosity young stellar objects ( above @xmath11 ) ,
this one is associated with an object of low luminocity ( @xmath12 ) .
slysh et al . ( 1999 ) showed that even quasi - thermal @xmath3 lines are typically inverted and their quasi - thermal appearance indicates that the line opacities are not large enough to cause significant narrowing . since the excitation of the @xmath0 transition is similar to that of the @xmath3 transition it is possible that the quasi - thermal @xmath0 lines are also inverted . to test this hypothesis , we determined the excitation temperature of the @xmath0 lines using the intensities of the @xmath13 lines at 157.2 ghz , measured by slysh et al .
the excitation temperatures were derived analytically using a simple method described by slysh et al .
we applied this method to 20 quasi - thermal sources , and for each , obtained negative excitation temperature between @xmath14 k and @xmath15 k , i.e. , the @xmath0 quasi - thermal lines proved to be strongly inverted .
the excitation temperatures derived in this way are distorted by a number of factors , such as the line opacities , influence of microwave background etc ( slysh et al . , 1999 ) .
therefore , we verified the results using a grid of lvg methanol models spanning the ranges @xmath16 @xmath17 in density , 10100 k in temperature , and @xmath18 @xmath17/(km / s pc@xmath19 ) in methanol density divided by the velocity gradient .
for each source , we selected the models corresponding to the observed ratios of the main - beam brightness temperatures of the @xmath0 line and the @xmath13 and @xmath20 lines , observed by slysh et al .
the results are as follows : for the majority of the sources , we found that only models with inversion of the @xmath0 transition or models with unlikely high methanol abundances satisfy the observed line ratios . in g29.95 - 0.02 , g34.26 + 0.15 ,
ngc 7538 , w 49n , and w 51e1/e2 , the observed intensity ratios can be obtained both in models with the inversion and in realistic models with positive excitation temperatures . however , since a number of models with inversion ( i.e. , same as those for the other 15 sources ) are applicable to these objects as well , it is not clear whether they are somehow different from the others or not .
thus , the quasi - thermal @xmath0 methanol lines , like the @xmath3 lines , are typically inverted .
this result confirms the plausibility of models in which compact class i masers appear in extended sources as a result of an appropriate velocity field ( see , e.g. , sobolev et al .
in the series of observations , performed in june 2000 with the 12-m nrao telescope at kitt - peak we tried to find linear polarization at 84.5 ghz towards 13 sources .
we expected that class i methanol masers may arise in a gas permeated by magnetic field and may exhibit a weak linear polarization similar to that of some h@xmath21o masers .
two polarization channels of the 3-mm receiver at kitt - peak can measure both senses of linear polarization simultaneously .
different brightness temperatures , measured in different channels would mean that the radiation is linearly polarized .
one can test whether the difference is a result of linear polarization by tracing the source during a sufficiently long time range .
owing to the diurnal rotation of the sky the direction of the polarization plane will vary , resulting in regular variations of the measured brightness temperatures in the two channels and hence , in a regular variation of the difference between them .
we failed to find any difference between channels in 12 sources and obtained the upper limits for the degree of linear polarization within the range 3%30% .
the only exception is a strong maser m 8e .
here we found a small difference between polarization channels , which might appear due to a weak ( @xmath22 ) linear polarization at 84.5 ghz .
unfortunately , this southern source ( @xmath23 ) can not be traced along a significant part of its diurnal trajectory from the northern hemisphere .
therefore , we could not detect any regular variation of the difference between channels and can not state that the difference is a result of linear polarization .
further polarization measurements of this source at 84.5 ghz are required .
the authors are grateful to the staff of the onsala space observatory and to the staff of the kitt peak telescope for providing help during the observations .
the work was done under a partial financial support of the russian foundation for basic research ( grant no . 95 - 02 - 05826 ) and intas ( grant no . 97 - 1451 ) .
batrla , w. & menten , k.m . , 1988 , , 329 , l117 kalenskii , s.v . ,
slysh , v.i . , valtts , i.e. , winnberg a. , & johansson l.e.b . , 2001 ,
astronomy reports , 45 , 26 menten k.m . , in : publ .
pac . , skylines , proc .
3rd haystack observatory meeting , eds .
haschick , p.t.p .
ho , 119 slysh v.i . , kalenskii s.v . , valtts i.e. , golubev v.v . , mead k. , 1999 , , 123 , 515 sobolev , a.m. , wallin , b.k . , & watson , w.d .
, 1998 , , 498 , 763 valtts , i.e. , dzura , a.m. , kalenskii , s.v . ,
slysh , v.i .
, booth r. , & winnberg , a. , 1995 , astronomy reports , 39 , 18 | fifty - one object in the @xmath0 methanol line at 84.5 ghz was detected during a survey of class i maser sources .
narrow maser features were found in 17 of these .
broad quasi - thermal lines were detected towards other sources .
one of the objects with narrow features , the young bipolar outflow l 1157 was also observed in the @xmath1 line at 95.2 ghz ; a narrow line was detected at this frequency .
analysis showed that the broad lines are usually inverted .
the quasi - thermal profiles imply that the line opacities are not larger than several units .
these results confirm the plausibility of models in which compact class i masers appear in extended sources as a result of an appropriate velocity field .
measurements of linear polarization at 84.5 ghz in 13 sources were made .
no polarization was found except a tentative detection of a weak polarization in m 8e .
# 1_#1 _ # 1_#1 _ = # 1 1.25 in .125 in .25 in |
the notion of spin currents describing the motion of magnetic moments associated with the spins of the electrons in solids is of central importance in the field of spintronics where one tries to use the spin degree of freedom to store and process information .
unfortunately , in systems lacking spin - rotational invariance ( which can be broken by an external magnetic field or by relativistic effects such as spin - orbit coupling or dipole - dipole interactions ) the proper definition of the quantum mechanical operator representing the spin current is ambiguous , because the magnetization does not satisfy a local conservation law . in the past decade
several authors have proposed resolutions of this ambiguity @xcite , but a generally accepted agreement on the correct definition of the spin current operator in systems without spin conservation has not been found . the purpose of this work is show that the distinction between translational and angular spin currents proposed by sun and xie @xcite leads to a simple and unique definition of the concept of spin transport in condensed matter systems .
sun and xie @xcite pointed out that spin currents describe moving magnetic dipoles , and that generally the transport of any vector can be decomposed into a translational part characterized by some velocity @xmath0 and an angular part described by some angular velocity @xmath1 , see fig .
[ fig : transport ] . in the context of spin transport sun and
xie @xcite called the latter contribution the _ angular spin current _ , although this can be also viewed as the spin torque discussed earlier by culcer _
et al . _
[ ] . with a velocity @xmath2 , corresponding to the physical movement of a magnetic moment with magnitude @xmath3 .
( b ) precessional motion of the magnetic polarization @xmath4 with a frequency @xmath5 .
the latter motion is purely angular and leaves the magnitude @xmath3 of the magnetic moment invariant .
, scaledwidth=40.0% ] we will explicitly show below that the equation of motion of a magnetic moment @xmath6 at lattice site @xmath7 with magnitude @xmath8 and polarization @xmath9 can be decomposed into a translational part @xmath10 that corresponds to the physical movement of magnetic dipole moments , and an angular part @xmath11 that describes the precessional motion of the magnetization .
we will provide expressions for the translational spin current operator @xmath12 and the precession frequency @xmath13 valid for itinerant as well as localized systems .
although the explicit construction of this decomposition is very simple , it entails profound physical consequences : since only the translational movement of magnetization corresponds to the physical displacement of magnetic moments , in equilibrium only stationary translational spin currents can generate an electrical field @xcite .
angular spin currents on the other hand only transport the polarization , hence a stationary angular spin current is simply an inhomogeneous configuration of the local magnetic order and does not create an electrical field .
furthermore , we also show that in generic magnetic insulators the classical magnetic order can not support translational spin transport at vanishing temperature ; incoherent thermal or quantum fluctuations are mandatory for the physical transport of magnetization in these systems .
this also implies that spin superfluidity in magnetic insulators @xcite must be angular spin transport that can be visualized as transporting the spin polarization , but does not correspond to the physical movement of magnetic moments .
the remainder of this work is organized as follows : in sec .
[ sec : separation ] we will derive general expressions for the operators corresponding to the translational spin current and to the precession frequency of the magnetization , first for itinerant systems and then for localized magnetic moments . we proceed to illustrate the usefulness of the decoupling procedure in sec . [ sec : supercurrents ] , where we discuss spin superfluidity in easy - plane ferromagnets and persistent spin currents in heisenberg rings . finally , in sec .
[ sec : conclusions ] we present our conclusions .
the appendix contains some additional details of the self - consistent spin - wave expansion we employ to describe easy - plane ferromagnets .
in this section , we explicitly show how translational and angular spin transport can be defined on the operator level . to construct the proper quantum mechanical definition of the translational spin transport operator
let us consider a lattice model describing electrons with spin - dependent hopping @xmath14 in an inhomogeneous magnetic field @xmath15 .
the second quantized hamiltonian of our model is @xmath16 where @xmath17 is some spin - rotationally invariant interaction , @xmath18 annihilates a fermion with spin - projection @xmath19 at lattice site @xmath20 , and the itinerant spin operators are defined by @xmath21 here @xmath22 is the vector of pauli matrices .
the spin - dependent hopping energies @xmath14 are of the form @xmath23 where the vectors @xmath24 are proportional to the strength of the spin - orbit coupling .
the hermiticity of the hamiltonian implies the symmetries @xmath25 and @xmath26 . using the canonical anticommutation relations @xmath27 and the fact that the interaction is spin - rotationally invariant , @xmath28 = 0 $ ]
, we obtain the heisenberg equation of motion for the itinerant spins , @xmath29 where we have defined the operator @xmath30 it is tempting to associate this operator with the spin current describing the transport of spin from lattice site @xmath20 to lattice site @xmath31 .
it turns out , however , that a certain part of this operator simply renormalizes the external magnetic field and therefore can not be associated with translational spin transport . to isolate this contribution and identify the angular part which renormalizes the precessional motion of the spins , we take the quantum mechanical expectation value of both sides of the equation of motion ( [ eq : eomit ] ) and obtain a formally exact equation of motion for the magnetic moments
@xmath32 , @xmath33 where the spin torque is defined by @xmath34 to identify the contribution responsible for translational spin transport , we further decompose the vector @xmath35 into a longitudinal and a transverse part , @xmath36 where @xmath37 is the local spin polarization and @xmath38 writing @xmath39 where @xmath40 is the induced magnetic field perpendicular to the direction of @xmath41 , we see that the transverse part @xmath42 renormalizes the external magnetic field .
the total angular frequency relevant for the precessional motion of the magnetic moments is @xmath43 the term @xmath44 can be called angular spin current @xcite or spin torque @xcite and should be added to the external torque @xmath45 acting on the magnetic moments .
the expectation value of the equation of motion ( [ eq : eomd ] ) can now be written as @xmath46 from this expression it is easy to show that the spin torque does not contribute to the time - evolution of the magnitude @xmath47 of the magnetic moments , which satisfies the equation of motion@xcite @xmath48 in contrast , the precessional motion of the spin polarization is governed solely by the spin torque , @xmath49 in summary , the renormalized precession frequency associated with the angular spin current is @xmath50 while the operator representing the translational spin current is @xmath51 note that time - dependent changes in the length of the magnetization are always accompanied by translational spin transport .
on the other hand , stationary translational spin currents are also possible if the length of the magnetization is constant @xcite .
the above expressions have been derived for a lattice model for itinerant electrons .
it is instructive to work out the explicit form of the rotation vector @xmath52 and the translational spin current operator @xmath53 for a localized spin model containing only the spin degrees of freedom . for simplicity ,
let us specify the interaction to the on - site hubbard interaction @xmath54 , where @xmath55 .
assuming @xmath56 and a half - filled lattice , we can use a canonical transformation @xcite to derive from eq .
( [ eq : hubbard ] ) an effective hamiltonian involving only spin @xmath57 operators @xmath58 acting on the reduced hilbert space of singly occupied lattice sites
. the effective spin hamiltonian can be written as@xcite @xmath59 where the spin - spin interaction tensor has three contributions , @xmath60 here the isotropic exchange coupling @xmath61 and the antisymmetric dzyaloshinskii - moriya vector @xmath62 are given by @xmath63 = \frac{8}{u } \textrm{re } \left [ t_{ij } \bm{\lambda}_{ij}^ * \right ] , \end{aligned}\ ] ] while @xmath64 is a symmetric tensor in spin space with matrix elements @xmath65 .\ ] ] the heisenberg equation of motion can be written as @xmath66 where the operator @xmath67 is the strong coupling limit of the operator @xmath68 defined in eq .
( [ eq : islattice ] ) in the reduced spin hilbert space . here
@xmath69 is a tensor in spin space with matrix elements given by eq .
( [ eq : jtens ] ) .
alternatively , eq . ( [ eq : istrong ] ) can be obtained directly from eq .
( [ eq : islattice ] ) via a canonical transformation @xcite . with the substitution
@xmath70 the expressions ( [ eq : omegageneral ] ) and ( [ eq : igeneral ] ) for the local precession frequency and the longitudinal spin transport operator remain valid , so that we obtain @xmath71 in the classical limit and at vanishing temperature the quantum mechanical expectation value @xmath72 can be factorized , @xmath73 consequently the expectation value of the longitudinal spin current operator vanishes identically in this limit .
this implies that for a generic magnetic insulator with a spin hamiltonian of the form of eq .
( [ eq : hspin ] ) incoherent thermal or quantum fluctuations are a necessary prerequisite for the transport of actual magnetization .
the local precession frequency reduces in the same limit to @xmath74 .
\label{eq : omegaires } \end{aligned}\ ] ] the last term in eq .
( [ eq : omegaires ] ) is proportional to @xmath75 and hence does not contribute to @xmath76 so that we may write @xmath77 , where the renormalization of the magnetic field is given by @xmath78 note that eq .
( [ eq : hex2 ] ) can also be obtained by means of a simple mean - field decoupling of the spin hamiltonian in eq .
( [ eq : hspin ] ) .
to illustrate the differences between translational and angular spin transport and how this affects the interpretation of spin supercurrents , it is instructive to consider simple model systems that can support translational and angular spin currents in equilibrium or in metastable states
. let us first consider an easy - plane ferromagnet described by the spin @xmath79 heisenberg hamiltonian @xmath80 on a simple cubic lattice , with exchange coupling @xmath81 for nearest neighbors only , and an easy - plane anisotropy @xmath82 .
this kind of systems has served in the literature as elementary example for spin superfluidity @xcite .
we will now calculate the local precession frequency ( [ eq : omegaspin ] ) and the translational spin current ( [ eq : ispinparallel ] ) for this system to leading order in an @xmath83 expansion . to facilitate this
we expand the spin operators in a local basis defined by the instantaneous direction of the spin polarization @xmath84 : @xmath85 here , @xmath86 and @xmath87 are unit vectors chosen such that @xmath88 form a right - handed basis at every lattice site . in this basis
we may now bosonize the spin operators by means of a holstein - primakoff ( hp ) transformation , [ eq : hp ] @xmath89 where the @xmath90 are canonical bose operators . especially note that since we self - consistently define the quantization axis as the direction of the local magnetization , @xmath84 , by definition the hp bosons can never condense @xcite .
this is completely analogous to the fact that in the superfluid phase of interacting bosons the bogoliubov quasi - particles , which are the goldstone modes associated with the spontaneous breaking of the u(1)-symmetry in the superfluid state , do not condense provided the condensate wave - function is self - consistently defined via the solution of the gross - pitaevskii equation . with the hp bosonization ( [ eq : hp ] )
, we find that the leading order contributions to the precession frequency and the spin current are of order @xmath79 .
explicitly , the local precession frequency ( [ eq : omegaspin ] ) becomes @xmath91 to this order , the polarization equation of motion ( [ eq : polarization ] ) recasts the landau - lifshitz equation of classical spin dynamics .
assuming that the magnetic texture @xmath92 varies only slowly in space , we can take the continuum limit .
the exchange contribution to the polarization equation of motion can then be identified with the divergence of the classical spin current , @xmath93 where the classical spin current is defined as @xmath94 here @xmath95 , and @xmath96 is the distance between nearest neighbors .
an exact equilibrium solution for the polarization is then given by @xmath97 $ ] , with a local phase satisfying @xmath98 .
the @xmath99 freedom associated with the choice of @xmath100 lies at the heart of the concept of spin superfluidity @xcite . out of equilibrium
we can make the general ansatz @xmath101 \nonumber\\[.2 cm ] & + \bm{e}_z \rho(\bm{r},t ) , \end{aligned}\ ] ] which is depicted graphically in fig .
[ fig : easy_plane ] . denotes the deviation from the ground state with the magnetization lying in the @xmath102-@xmath103-plane .
superfluid angular transport is possible due to the @xmath99 freedom of the phase @xmath104 in this plane .
, scaledwidth=40.0% ] if the system is only slightly driven out of equilibrium , the equation of motion ( [ eq : polarization ] ) for the spin polarization becomes to lowest nonvanishing order in deviations from equilibrium [ eq : josephson ] @xmath105 the above eqs .
( [ eq : josephson ] ) are of the form of the josephson equations of superconductivity ; hence they are conventionally interpreted as describing a spin supercurrent carried by the magnetization texture @xcite . however , since this current does not correspond to translational transport of the local magnetization @xmath106 , this supercurrent is _ not _ equivalent to the superfluid transport of magnetic moments . in particular , a stationary angular supercurrent with @xmath107 is simply an inhomogeneous magnetic texture ; while it can be viewed as a stationary transport of polarization , it does not correspond to any physical movement of magnetic moments which would generate an electrical field @xcite . to evaluate the @xmath108 contribution to the translational spin current ( [ eq : ispinparallel ] )
, we will assume for simplicity a slowly varying magnetic texture @xmath75 in a metastable superfluid state with @xmath109 .
applying the hp transformation ( [ eq : hp ] ) to the spin current ( [ eq : ispinparallel ] ) then yields @xmath110 , \ ] ] which can to be evaluated to @xmath111 , \ ] ] where @xmath7 and @xmath112 are nearest neighbor lattice sites . here
@xmath113 is the distribution functions of magnons with dispersion @xmath114 and velocity @xmath115 , and the remaining coefficient of the quantum - mechanical zero - point fluctuations is @xmath116 a detailed derivation of the hamiltonian in the local basis and of the magnon dispersion ( [ eq : ek ] ) is relegated to the appendix . in equilibrium
the translational current ( [ eq : iplane ] ) of course vanishes by symmetry , i.e. , @xmath117 .
therefore there is no translational movement of magnetic moment in this superfluid spin state .
lastly , let us note that the complete decoupling of the magnetic texture and the incoherent magnons is an artifact of the lowest order approximation in the @xmath83 expansion , and of the assumption of a slowly varying texture .
if one relaxes either of these approximations , there will be a coupling , resulting in a two - fluid description of spin transport like the one discussed in [ ] .
we emphasize however that from the point of view we adopted , the superfluid does not arise due to the condensation of magnons .
we rather consider the magnons as fluctuations on top of the superfluid ground state ; these magnons can not condense by definition .
next , let us consider a system where the translational spin current is finite even in equilibrium , i.e. , there is a persistent translational spin current corresponding to the phyiscal movement of magnetic moments .
consider a ferromagnetic spin @xmath79 heisenberg model in a radial inhomogeneous magnetic field with quantum mechanical hamiltonian @xmath118 where the sums are over the @xmath119 sites of the lattice of localized spins on the ring coupled by ferromagnetic exchange interactions @xmath120 if @xmath121 and @xmath122 label nearest neighbors .
@xmath15 is a crown - shaped inhomogeneous magnetic field of the form @xmath123,\ ] ] where the angles @xmath124 label the positions of the spins on the ring , as illustrated in fig .
[ fig : crown ] . between the magnetization and
the @xmath125-axis is slightly smaller then the angle @xmath126 between the external field and the @xmath125-axis , see ref .
[ ] . , scaledwidth=40.0% ] in ref .
[ ] it has been shown that at finite temperature @xmath127 spin - wave excitations carry a persistent equilibrium spin current @xmath128 circulating the ring . here
@xmath129 is the length of the ring , @xmath130 is the magnon velocity , and the magnon dispersion is @xmath131 , with lattice spacing @xmath96 and quantized wavevectors @xmath132 .
@xmath133 is the solid angle traced out by the local magnetization direction @xmath75 on the unit sphere in order parameter space as it moves around the chain .
this finite solid angle @xmath133 , i.e. , the topology of the spin configuration on the ring , is responsible for the finiteness of the equilibrium current ( [ eq : iring ] ) .
the situation is completely analogous to persistent electrical currents in mesoscopic metal rings pierced by a magnetic flux @xcite .
these electrical currents generate a magnetic dipole field ; one of us has shown in ref . [ ]
that the persistent spin current ( [ eq : iring ] ) similarly generates an electric dipole field .
this can be understood as follows @xcite : a magnetic dipole moment @xmath134 moving with velocity @xmath2 generates a magnetic field @xmath135 in its rest frame . lorentz - transforming back to the laboratory frame , we find that to lowest order in @xmath136 ( where @xmath137 is the speed of light ) this magnetic field generates an electrical field @xmath138 . at zero temperature the equilibrium current ( [ eq : iring ] ) vanishes because there are no spin waves in the ferromagnetic ground state . however , bruno and dugaev @xcite pointed out that in this system the classical spin current @xmath139 defined in eq .
( [ eq : jclassical ] ) is finite and argued that therefore the system exhibits an equilibrium spin supercurrent even at @xmath140 .
while this interpretation is possible , we stress that this classical current is angular and not translational ; hence it can not be associated with the motion of magnetic dipoles , but should rather be considered as a renormalization of the external magnetic field @xcite . to understand this ,
let us explicitly calculate the inhomogeneous magnetization configuration in the classical ground state of the hamiltonian ( [ eq : hamiltoniancrown ] ) .
just as in the easy - plane ferromagnet , the equation of motion for the magnetic texture @xmath75 is to leading order in @xmath83 the classical landau lifshitz equation @xmath141 where the exchange field is @xmath142 .
the equilibrium solution of the landau - lifshitz equation ( [ eq : llring ] ) is @xmath143 where the angle @xmath144 is slightly smaller than the angle @xmath126 between the magnetic field and the @xmath125-axis , as discussed in ref .
the deviation of @xmath144 from @xmath126 is determined by the exchange field @xmath145 . as for the easy - plane ferromagnet discussed in the last section , the continuum limit of the exchange torque
is the divergence of the classical spin current , @xmath146 . for the classical ground state ( [ eq : groundstatering ] ) , the classical spin current has a finite component @xmath147 .
however , as in the easy - plane ferromagnet , this finite current merely signals an inhomogeneous magnetization configuration and is not related to the physical transport of magnetization ; hence it also will not generate any electrical field .
if one would on the other hand incorrectly associate the classical spin current @xmath148 with the stationary movement of physical dipoles , it would have to be accompanied by an electrical field .
this would imply that a purely static inhomogeneous magnetization configuration is always accompanied by an electric field , in contradiction with the elementary fact that in classical electromagnetism magnetostatics and electrostatics are completely decoupled .
finally , let us also point out that although the classical ground state ( [ eq : groundstatering ] ) of the heisenberg ring is very similar to the classical ground state of the easy - plane ferromagnet discussed in the last section , it does not support spin superfluidity , i.e. , angular spin supercurrents , because the phase @xmath124 of the magnetization is pinned by the external magnetic field @xmath149 , eq .
( [ eq : hcrown ] ) .
we conclude that in the classical limit the angular spin current introduced by sun and xie @xcite can be absorbed into a renormalization of the external magnetic field which contributes to the torque acting on the magnetic moments . in equilibrium
, this term does not describe any current of magnetic moments because the equilibrium configuration of the magnetization is such that the total torque on each moment vanishes .
translational transport of spins is described by the longitudinal spin current defined in eq .
( [ eq : ispinparallel ] ) , which is only finite due to thermal or quantum fluctuations .
an example for a system exhibiting a finite longitudinal spin current in equilibrium is a mesoscopic heisenberg ring in a crown - shaped magnetic field , as discussed in ref .
our considerations imply that in equilibrium the classical spin current @xmath150 defined in eq .
( [ eq : jclassical ] ) does not describe any motion of magnetization . while it can be interpreted as a stationary current of magnetic polarization ,
such an interpretation is by no means mandatory since there is no phyiscal movement . in nonequilibrium on the other hand
, the classical spin current @xmath150 contributes to the angular spin current , i.e. , to the precessional motion of the spins , and does transport spin polarization . in particular
, this implies that spin superfluidity , which is based on the formal similarity of eq .
( [ eq : josephson ] ) with a mass supercurrent of superfluid bosons @xcite , does not correspond to the physical movement of _ magnetization _ , but of _ polarization_.
this means that in equilibrium , a superfluid spin state will not be accompanied by an electrical field , in contrast to a persistent translational spin current @xcite .
this physical difference between the two types of spin transport persists also out of equilibrium : as already shown by sun and xie @xcite , an angular spin current with finite @xmath151 will generate an electrical field @xmath152 for large distance @xmath153 from the source , whereas the electrical field of translational spin currents decays as @xmath154 .
we have also shown that spin superfluidity can be described entirely without referring to off - diagonal long - range order @xcite and magnon condensation .
this is achieved by quantizing the spins in a self - consistently defined frame of reference with the local @xmath125-axis pointing in the direction @xmath155 of the instantaneous magnetization .
magnons defined with respect to this reference frame can never condense or display off - diagonal long - range order , hence they are not superfluid .
this is in agreement with the general proof of kohn and sherrington @xcite that bosonic quasi - particles which are formed as bound states of particle - hole pairs of the underlying fermionic system ( such as excitons or magnons ) do not exhibit off - diagonal long - range order in coordinate space .
hence , the bose - einstein condensation of this type of bosons is not accompanied by superfluidity .
although the change in magnetic order in a magnetic insulator can be viewed as bose - einstein condensation of magnons @xcite , the resulting state can always be characterized by magnons that exhibit neither off - diagonal long - range order nor superfluidity .
this work was financially supported by the dfg via sfb / trr49 .
our ideas on spin transport were sharpened during a workshop on _ quantum spintronics : spin transport through quantum magnetic materials _ at the spin phenomena interdisciplinary center ( spice ) at the university of mainz , germany .
this appendix is devoted to the derivation of the dispersion ( [ eq : ek ] ) of the magnons in the local reference frame defined by the magnetic polarization @xmath84 .
we have already derived the general setup of a spin - wave expansion in this local and possibly time dependent reference frame in ref .
the first step is to rotate the @xmath125-axis of the laboratory frame to the direction @xmath156 of the local magnetization by means of an unitary transformation @xmath157 acting on the spins .
the explicit form of @xmath157 is given in ref .
it then turns out that the rotated hamiltonian @xmath158 contains an additional berry - phase term acting as a magnetic field due to the time dependence of @xmath156 , @xmath159 writing the spin polarization as @xmath160 + \bm{e}_z \cos\theta_i , \ ] ] we can choose the transverse basis vectors as @xmath161 + \bm{e}_z \sin\theta_i . \ ] ] expanding the berry - phase magnetic field in this basis , @xmath162 , we explicitly find applying the holstein - primakoff transformation ( [ eq : hp ] ) to the rotated hamiltonian then yields @xmath164 here the classical ground state energy is @xmath165 the term linear in the bose operators can be written as @xmath166 where @xmath13 is the @xmath108 local precession frequency , eq .
( [ eq : omegas ] ) .
because the magnetic polarization satisfies the equation of motion @xmath167 up to this order in @xmath83 , we conclude that @xmath168 vanishes identically .
note that this implies that the holstein - primakoff bosons can not condense , a statement which remains true to all orders in @xmath83 due to the self - consistency of the basis .
up to this point the discussion is completely general and applies to all magnetic insulators . to evaluate the quadratic part of the rotated hamiltonian we will assume as in the main text that the magnetic texture varies only slowly in space and is only slightly out of equilibrium .
then we may approximate @xmath169 , @xmath170 , @xmath171 , with @xmath172 , and @xmath173 .
with this simplifications we immediately obtain @xmath174 , \end{aligned}\ ] ] which describes free magnons with the dispersion ( [ eq : ek ] ) .
99 e. i. rashba , https://journals.aps.org/prb/abstract/10.1103/physrevb.68.241315[_spin currents in thermodynamic equilibrium : the challenge of discerning transport currents _ ] , phys .
b * 68 * , 241315(r ) ( 2003 ) .
e. i. rashba , https://journals.aps.org/prb/abstract/10.1103/physrevb.70.161201[_spin currents , spin populations , and dielectric function of noncentrosymmetric semiconductors _ ] , phys .
b * 70 * , 161201 ( 2004 ) .
e. i. rashba , http://link.springer.com/article/10.1007/s10948-005-3349-8[_spin dynamics and spin transport _ ] , j. supercond .
* 18 * , 137 ( 2005 ) .
f. schtz , m. kollar , and p. kopietz , https://journals.aps.org/prl/abstract/10.1103/physrevlett.91.017205[_persistent spin currents in mesoscopic heisenberg rings _ ] , phys .
lett . * 91 * , 017205 ( 2003 ) .
f. schtz , p. kopietz , and m. kollar , http://link.springer.com/article/10.1140%2fepjb%2fe2004-00348-9[_what are spin currents in heisenberg magnets ? _ ] , eur .
j. b * 41 * , 557 ( 2004 ) .
d. culcer , j. sinova , n. a. sinistyn , t. jungwirth , a. h. macdonald , and q. niu , https://journals.aps.org/prl/abstract/10.1103/physrevlett.93.046602[_semiclassical spin transport in spin - orbit - coupled bands _ ] , phys .
93 * , 046602 ( 2004 ) .
p. bruno and v. k. dugaev , https://journals.aps.org/prb/abstract/10.1103/physrevb.72.241302[_equilibrium spin currents and the magnetoelectric effect in magnetic nanostructures _ ] , phys .
b * 72 * , 241302(r ) ( 2005 ) . q .-
f . sun and x. c. xie , https://journals.aps.org/prb/abstract/10.1103/physrevb.72.245305[_definition of the spin current : the angular spin current and its physical consequences _ ] , phys .
b * 72 * , 245305 ( 2005 ) .
j. shi , p. zhang , d. xiao , and q. niu , https://journals.aps.org/prl/abstract/10.1103/physrevlett.96.076604[_proper definition of spin current in spin - orbit coupled systems _ ] , phys .
* 96 * , 076604 ( 2006 ) .
p. zhang , z. wang , j. shi , d. xiao , and q. niu , https://journals.aps.org/prb/abstract/10.1103/physrevb.77.075304[_theory of conserved spin current and its application to a two - dimensional hole gas _ ] , phys . rev .
b * 77 * , 075304 ( 2008 ) .
i. g. bostrem , jun - ichiro kishine , and a. s. ovchinnikov , https://journals.aps.org/prb/abstract/10.1103/physrevb.78.064425[_theory of spin current in chiral helimagnets _ ] , phys . rev . * 78 * , 064425 ( 2008 ) .
i. v. tokatly , https://journals.aps.org/prl/abstract/10.1103/physrevlett.101.106601[_equilibrium spin currents : non - abelian gauge invariance and color diamagnetism in condensed matter _ ] , phys .
* 101 * , 106601 ( 2008 ) .
n. bray - ali and z. nussinov , https://journals.aps.org/prb/abstract/10.1103/physrevb.80.012401[_conservation and persistence of spin currents and their relation to the lieb - schulz - mattis twist operators _ ] , phys . rev .
b * 80 * , 012401 ( 2009 ) .
z. an , f. q. liu , y. lin , and c. liu , http://www.nature.com/articles/srep00388[_the universal definition of spin current _ ] , sci .
* 2 * , 388 ( 2012 ) .
b. berche , e. medina , and a. lpez , http://iopscience.iop.org/article/10.1209/0295-5075/97/67007/meta[_spin superfluidity and spin - orbit gauge symmetry fixing _ ] , europhys . lett . * 97 * , 67007 ( 2012 ) . k. nakata , k. a. van hoogdalem , p. simon , and d. loss , https://journals.aps.org/prb/abstract/10.1103/physrevb.90.144419[_josephson and persistent spin currents in bose - einstein condensates of magnons _ ] , phys .
b * 90 * , 144419 ( 2014 ) .
j. e. hirsch , https://journals.aps.org/prb/abstract/10.1103/physrevb.60.14787[_overlooked contribution to the hall effect in ferromagnetic metals _ ] , phys .
b * 60 * , 14787 ( 1999 ) .
j. knig , m. chr .
bnsager , and a. h. macdonald , https://journals.aps.org/prl/abstract/10.1103/physrevlett.87.187202[_dissipationless spin transport in thin film ferromagnets _ ] , phys .
rev . lett . * 87 * , 187202 ( 2001 ) .
f. s. nogueira , k .- h . bennemann , _ spin josephson effect in ferromagnet / ferromagnet tunnel junctions _ , europhys .
* 67 * , 620 ( 2004 ) .
e. b. sonin , _ spin currents and spin superfluidity _ , adv
. phys . * 59 * , 181 ( 2010 ) .
e. b. sonin , _ spin superfluidity , coherent spin precession , and magnon bec _ , j. low temp
. phys . * 171 * , 757 ( 2013 ) .
h. chen , a. d. kent , a. h. macdonald , and i. sodemann , https://journals.aps.org/prb/abstract/10.1103/physrevb.90.220401[_nonlocal transport mediated by spin supercurrents _ ] , phys .
b * 90 * , 220401(r ) ( 2014 ) .
s. takei and y. tserkovnyak , https://journals.aps.org/prl/abstract/10.1103/physrevlett.112.227201[_superfluid spin transport through easy - plane ferromagnetic insulators _ ] , phys .
lett . * 112 * , 227201 ( 2014 ) .
s. takei , b. i. halperin , a. yacoby , and y. tserkovnyak , https://journals.aps.org/prb/abstract/10.1103/physrevb.90.094408[_superfluid spin transport through antiferromagnetic insulators _ ] , phys .
b * 90 * , 094408 ( 2014 ) .
b. flebus , s. a. bender , y. tserkovnyak , and r. a. duine , https://journals.aps.org/prl/abstract/10.1103/physrevlett.116.117201[_two-fluid theory for spin superfluidity in magnetic insulators _ ] , phys .
116 * , 117201 ( 2016 ) .
h. chen and a. h. macdonald , http://de.arxiv.org/abs/1604.02429[_spin-superfluidity and spin - current mediated non - local transport _ ] , arxiv:1604.02429 .
a. h. macdonald , s. m. girvin , and d. yoshioka , https://journals.aps.org/prb/abstract/10.1103/physrevb.37.9753[_t/u expansion for the hubbard model _ ] , phys .
b * 37 * , 9753 ( 1988 ) .
t. moriya , https://journals.aps.org/pr/abstract/10.1103/physrev.120.91[_anisotropic superexchange interaction and weak ferromagnetism _ ] , phys .
rev . * 120 * , 91 ( 1960 ) .
a. rckriegel , a. kreisel , and p. kopietz , https://journals.aps.org/prb/abstract/10.1103/physrevb.85.054422[_time-dependent spin - wave theory _ ] , phys .
b * 85 * , 054422 ( 2012 ) . c. n. yang , https://journals.aps.org/rmp/abstract/10.1103/revmodphys.34.694[_concept of off - diagonal long - range order and the quantum phases of liquid he and of superconductors _ ] , rev
phys . * 34 * , 694 ( 1962 ) .
w. kohn and d. sherrington , https://journals.aps.org/rmp/abstract/10.1103/revmodphys.42.1[_two kinds of bosons and bose condensates _ ] , rev .
phys . * 42 * , 1 ( 1970 ) .
s. o. demokritov , v. e. demidov , o. dzyapko , g. a. melkov , a. a. serga , b. hillebrands , and a. n. slavin , http://www.nature.com/nature/journal/v443/n7110/full/nature05117.html[_boseeinstein condensation of quasi - equilibrium magnons at room temperature under pumping _ ] , nature ( london ) * 443 * , 430 ( 2006 ) .
a. rckriegel and p. kopietz , https://journals.aps.org/prl/abstract/10.1103/physrevlett.115.157203[_rayleigh-jeans condensation of pumped magnons in thin - film ferromagnets _ ] , phys .
lett . * 115 * 157203 ( 2015 ) . | in magnets with non - collinear spin configuration the expectation value of the conventionally defined spin current operator contains a contribution which renormalizes an external magnetic field and hence affects only the precessional motion of the spin polarization .
this term , which has been named angular spin current by sun and xie [ phys .
rev b * 72 * , 245305 ( 2005 ) ] , does not describe the translational motion of magnetic moments .
we give a prescription how to separate these two types of spin transport and show that the translational movement of the spin is always polarized along the direction of the local magnetization .
we also show that at vanishing temperature the classical magnetic order parameter in magnetic insulators can not carry a translational spin current , and elucidate how this affects the interpretation of spin supercurrents .
= 10000 |
the past 15 years have seen large advances in the capabilities of ground - based gamma ray detection , from the pioneering observation of the crab nebula by the whipple observatory in 1989@xcite to the new generation of air cherenkov telescope arrays such as hess@xcite , veritas@xcite , and cangaroo@xcite and large area air cherenkov telescopes such as stacee @xcite , celeste@xcite , and magic@xcite .
there are now at least 10 known sources of very - high - energy ( vhe ) gamma rays@xcite .
the physics of these objects is astounding : from spinning neutron stars to super - massive black holes , these objects manage to accelerate particles to energies well in excess of 10 tev .
how this acceleration occurs is not well understood and there is not universal agreement on what particles are being accelerated in some of these sources . at lower energies egret
has detected over 270 sources of high - energy gamma rays@xcite and glast is expected to detect several thousand sources .
in addition there are transient sources such as gamma - ray bursts that have to date eluded conclusive detection in the vhe regime ( despite some tantalizing hints@xcite ) .
the paucity of vhe sources can be traced to the nature of the existing instruments : they are either narrow field instruments that can only view a small region of the sky at any one time and can only operate on clear moonless nights ( whipple , hegra , etc . ) or large field instruments with limited sensitivity ( milagro , tibet array ) .
the milagro observatory has pioneered the use of a large area water cherenkov detector for the detection of extensive air showers . since an extensive air shower ( eas )
array directly detects the particles that survive to ground level it can operate continuously and simultaneously view the entire overhead sky . with the observation of the crab nebula and the active galaxies mrk 421 and mrk 501 , milagro has proven the efficacy of the technique and its ability to reject the cosmic - ray background at a level sufficient to detect sources@xcite . at the same time
the tibet group@xcite has demonstrated the importance of a high - altitude site and what can be accomplished with a classical scintillator array at extreme altitudes . a detector with the all - sky and high - duty factor capabilities of milagro , but with a substantially lower energy threshold and a greatly increased sensitivity ,
would dramatically improve our knowledge of the vhe universe .
reasonable design goals for such an instrument are : * ability to detect gamma - ray bursts to a redshift of 1.0 * ability to detect agn to a redshift beyond 0.3 * ability to resolve agn flares at the intensities and durations observed by the current generation of acts * ability to detect the crab nebula in a single transit this paper describes a design for a next generation all - sky vhe gamma - ray telescope , the hawc ( high altitude water cherenkov ) array , that satisfies these requirements . to quantify the definition of observing `` short '' flares from agn ,
previous measurements of flare intensities and durations by air cherenkov telescopes can be used .
to date the shortest observed flares have had @xmath015 minute durations with an intensity of 3 - 4 times that of the crab@xcite .
the low energy threshold needed to accomplish these goals requires that the detector be placed at extreme altitudes ( hawc would be situated at an altitude of @xmath04500 meters ) and the required sensitivity demands a large area detector - of order 40,000 m@xmath2 .
section [ sec : particle_detection ] discusses the limiting performance of an eas array based on the properties of the eas , section [ sec : detector_description ] gives a physical description of the hawc and section [ sec : detector_performance ] details the expected performance of hawc .
the ultimate performance of an eas array will be determined by the number , type , position , arrival time , and energy of the particles that reach the ground . here
these properties of air showers are investigated to arrive at the limiting performance of eas arrays . to attain this level of performance
an eas array would have to measure each of the above parameters with good precision .
the most well - studied aspect of eas is the dependence of the number of particles to reach ground level on the observation altitude . for electromagnetic cascades ,
approximation b is a good estimator of the average number of particles in an eas as a function of atmospheric depth .
however , at the threshold of an eas array , it is the fluctuations in the shower development that determines the response of the detector . to incorporate the effect of shower fluctuations the event generator corsika ( version 6.003@xcite )
is used to generate eas from gamma rays .
the gamma rays were generated with an @xmath3 spectrum beginning at 10 gev , and uniformly over over the sky with zenith angles from 0 to 45 degrees .
four different observation altitudes were studied : 2500 m , 3500 m , 4500 m , and 5200 m .
figure [ fig : f1-altitude_effect ] shows the fraction of primary gamma rays that generated an air shower where more than 100 particles with energy above 10 mev survived to the observation level .
the requirement that a particle have at least 10 mev in energy is imposed as a reasonable detection criterion and the requirement that 100 such particles survive enables one to examine the effects of altitude on the fluctuation spectrum of the air showers .
this figure is a reasonable indication of the relative effective area of a fixed detector design as a function of the altitude of the detector . at high energies
each km in altitude results in a factor of 2 - 3 increase in effective area . at low energies ( of relevance for extragalactic sources such as grbs ) the increase with altitude
is larger .
note that for primary energies between 100 gev and 500 gev a detector placed at 5200 m has @xmath4 times more effective area than the identical detector placed at 2500 m altitude ( close to the altitude of milagro ) . from this figure alone
one can estimate that a detector roughly 10 times the size of milagro placed at 5200 m altitude would be of order 22 ( @xmath5 ) times more sensitive than milagro ( assuming the background scaled in a similar fashion ) .
this level of sensitivity satisfies the requirements listed above .
-0.3 in , title="fig : " ] it is well known that gamma rays outnumber electrons and positrons in an air shower by a large factor .
figure [ fig : gamma - electron_ratio ] shows the ratio of gamma rays to electrons ( @xmath110 mev ) for three observation altitudes : 3500 m , 4500 m , and 5200 m . over this altitude range the ratio of gamma rays to electrons is relatively independent of the observation level . at these low energies gamma rays outnumber electrons and positrons by an order of magnitude . therefore to attain the lowest possible energy threshold the detector must be sensitive to the gamma rays in the air shower
. an important feature of any gamma - ray detector is its ability to reject a significant fraction of the cosmic - ray background .
the milagro detector has demonstrated that detecting and characterizing the penetrating component of an air shower is a reliable method of rejecting the cosmic - ray background@xcite . in figure [ fig : hadron_muon ] we show the average number of hadrons and muons in an air shower as a function of primary proton energy .
an observation altitude of 5200 m is assumed in this figure .
it should be noted that the fluctuations about these mean values are * not * poisson .
but it is clear from this figure that the ability to detect showering hadrons in the air shower is important for the discrimination of the cosmic - ray background .
this is especially true at low energies where the hadronic component is larger than the muon component .
( hadrons are defined to be protons , anti - protons , neutrons , and pions . )
-0.3 in , title="fig : " ] -0.3 in , title="fig : " ] after an event is recorded its direction in the sky must be reconstructed . to establish a limiting angular resolution for an eas array gamma - ray induced eas
are generated using corsika as above ( on an @xmath3 spectrum beginning at 10 gev ) , however only vertical showers were generated .
this simplifies the reconstruction of the air shower .
in what follows it is assumed that the position , time , and energy of each particle are measured exactly . since the direction of the individual particle is not utilized in what follows it may be possible to develop a detector with improved angular resolution , though at a rather large expense .
it is beyond the scope of this paper to give a detailed description of the fitting procedure .
the leading edge of the eas is curved : particles farther from the core of the air shower arrive late relative to the particles at the core .
studies of the corsika showers show that this curvature is dependent upon the individual energies of the shower particles .
lower energy particles are delayed longer than higher energy particles . before using the arrival times of the particles to fit the direction of the primary gamma ray the particle arrival times
must be corrected for this effect . to demonstrate the size of this effect figure [ fig : particle - point - spread ] shows the point spread function before and after this correction
is applied to the particle arrival times ( for an observation altitude of 5200 m ) . the median angle error before the corrections are applied is 5.1 degrees and after the corrections are applied the median angle error is reduced to 0.8 degrees ( 71% of the events survive the complete fitting procedure ) .
figure [ fig : particle - angle_comparealtitude ] shows the energy dependence of the mean of the point - spread function as a function of primary energy . for a fixed primary energy the angular reconstruction is improved by moving the detector to higher altitude .
a detector at 4500 m would have the same angular resolution as a detector at 5200 m for gamma - ray primaries with 50% more energy .
a detector at 3500 m altitude would require over a 100% increase in primary energy to obtain a similar angular resolution .
-0.3 in , title="fig : " ] -0.3 in , title="fig : " ] because of the fluctuations in the shower development ( mainly the depth of the first interaction ) eas arrays have relatively poor energy resolution . to investigate the limiting energy resolution of an eas array primary gamma rays
were generated on an @xmath3 spectrum from the zenith and the total number of particles with more than 10 mev that reach the ground within 100 m of the shower core were counted .
figure [ fig : eres - particles_compare ] shows the mean ( and the error on the mean ) number of particles on the ground as a function of primary energy for observation altitudes of 3500 m , 4500 m , and 5200 m . at an altitude of 5200 m there is a good correlation between the mean number of particles and primary energy at primary energies as low as 50 gev . at an altitude of 4500 m
this correlation begins at about 100 gev in primary energy and at 3500 m altitude there is no correlation until the primary energy is about 250 gev . to estimate the energy resolution , the curves in figure [ fig : eres - particles_compare ] are fit to a power law in the number of particles that reach the ground ( @xmath6 , where @xmath7 is the number of particles on the ground ) .
figure [ fig : particle_eres_gt_50gev ] shows the distribution of @xmath8 for primary energies above 50 gev ( for an observation altitude of 5200 m ) .
the resulting distribution is non - gaussian , but is fit reasonably well by a landau distribution ( shown on the figure ) .
the asymmetric tail is an over - estimation of the particle energy and is inherent in the nature of the fluctuations in the depth of the first interaction . with a full width at half maximum of 0.5
the equivalent size of the 1-sigma error bar would be @xmath025% ( if the distribution were gaussian ) .
as one goes to higher energy primaries the distribution becomes narrower and more gaussian . at energies below 50 gev
the energy resolution is significantly worse , with an rms greater than 1 .
since the fit is to the mean energy and there is a large positive tail , the peak in the distribution is negative .
-0.3 in , title="fig : " ] -0.3 in , title="fig : " ] this section concludes with a brief summary of the best performance that can be expected from an extensive air shower array operating at high altitude .
* an energy threshold below 50 gev * an angular resolution of 0.25 degrees at median energy * an energy resolution @xmath025% above 50 gev ( with a non - gaussian tail )
based on the above study and experience with the milagro detector a complete monte carlo simulation of a detector that is similar to milagro has been performed .
the need to detect the gamma rays in the eas and the penetrating component of muons and hadronic cascades leads to a thick active detector .
water is a natural and inexpensive choice of detecting medium .
the cherenkov photons are relatively plentiful , prompt , and have large mean free paths ( relative to the water depth ) in clean water . since the cherenkov angle in water is 41@xmath9 , a detector spacing of roughly twice the detector depth ensures that there are no geometrical blind areas .
( the depth and spacing were chosen to match that of milagro , though more work is needed to optimize these parameters . )
the choice of a two layer design was again driven by experience with milagro .
the deep layer makes a good calorimeter , whereas a shallow layer does not . in the shallow layer
the detected light level depends critically on the exact geometry of the incoming particles , their distance from the photomultiplier tube ( pmt ) and incident angle .
these fluctuations caused by the geometry of the incoming particles dilutes the energy measuring capability of a shallow layer of pmts .
the ability to measure the energy of the shower particles is needed to determine the penetrating component of the air shower . in milagro
this is the most useful method of rejecting the abundant cosmic - ray background .
a shallow layer may be required to obtain the best possible angular resolution ( also important for rejecting the cosmic - ray background ) . since the speed of light in water is substantially different from the speed at which the particles propagate ( the speed of light in vacuum ) , the light and the particles diverge .
this leads to an increase in the dispersion of the arrival time of the measured shower front as the depth of the measuring device is increased and the angular resolution is proportional to this spread .
this effect is somewhat compensated for by the fact that a deeper layer typically makes more measurements of the shower front ( because each pmt can see a larger area of the water surface ) .
the top layer may also be useful in the rejection of the cosmic ray background ( described below ) .
work in better understanding these effects is still underway , but it may be possible to eliminate the shallow layer and maintain the performance of the detector . in the remainder of this paper , the following detector characteristics are used .
* 40,000m@xmath2 physical area ( 200m@xmath10200 m ) * 2.7 m spacing of pmts ( 74@xmath1074 pmts per layer ) * hamamatsu 20 cm pmts ( r5912sel ) * two layers of pmts . a shallow layer under 1.5 m of water and a deep layer under 6.5 m of water * an attenuation length of 20 meters , composed solely of absorption .
* trigger requires 50 pmts in the top layer to be struck by at least 1 photo - electron ( pe ) * a detector latitude of 36 degree ( that of milagro ) corsika 6.003@xcite was used to generate the air showers and geant 3.21 was used to track the particles through the water , generate the cherenkov light in the water , and to model the detector response ( pmts , reflectivity of the material , scattering and absorption of light in water , etc . ) . for the remainder of this paper this design
will be referred to as the high altitude water cherenkov ( hawc ) array .
to obtain realistic estimates of the sensitivity of hawc , the individual events must be reconstructed - their core position , direction , type ( hadronic or gamma ray ) , and energy need to be determined .
preliminary algorithms have been developed to determine all of these parameters with the exception of the primary energy .
work is ongoing in improving the existing algorithms and developing one to determine the energy of the primary gamma ray . in
what follows , the events were generated beginning at 10 gev on an e@xmath11 spectrum ( similar to that observed for the crab nebula@xcite ) uniformly over incident direction . as discussed above ,
the shower front of an eas is not a true plane .
therefore , the first step in performing an angular reconstruction is the determination of the core of the air shower .
a simple center - of - mass algorithm using the pulse heights recorded in the pmts in the bottom layer of the detector is used to determine the shower core .
figure [ fig : hawc - core_resolution4572 ] shows the distribution of core reconstruction errors in meters . for all events that trigger the detector and are successfully reconstructed ( see below ) the mean core error is 36 meters . despite the simplicity of the algorithm the core resolution is good enough that it does not degrade the angular resolution .
the events with large core errors are those whose cores fall outside of the detector .
the dashed line in the figure shows the core error for events whose core falls within the detector , the mean of this distribution is 20 meters .
-0.3 in , title="fig : " ] the angular reconstruction is performed in an analogous manner to that described above , however the timing correction that was applied based on the individual particle energies is replaced by a correction based upon the number of pes measured in each pmt . after applying this correction and a correction for the shower front curvature ( proportional to the distance from the pmt to the core of the shower ) the resultant times and positions to a plane .
in addition the weight associated with each pmt is determined by the measured pulse height . though the weight also depends upon the distance to the shower core , this effect has been ignored in the following analysis .
a 20 - 30% improvement in the angular resolution may be expected with an improved algorithm .
this improvement should arise when the correlations between the two corrections ( core distance and pulse height ) are utilized and the weights assigned to each pmt incorporate the core distance .
the fitting procedure is iterative , where the pmts that made large contributions to the chi - square are removed in subsequent iterations .
events with more than 20 pmts surviving in the final iteration of the fitting procedure are considered to be successfully fit .
the space angle difference between the true direction and the fit direction is shown in figure [ fig : hawc - angular_resolution4572 ] . using this point - spread function the signal to noise ratio is maximal for an analysis bin size of 1.2 degrees .
if the point - spread function were gaussian this bin size would correspond to an angular resolution of 0.75 degrees@xcite .
-0.3 in , title="fig : " ] the ability to discriminate between gamma - ray induced air showers and hadronic air showers is critical to the success of any gamma - ray detector . unlike air cherenkov telescopes the background rejection discussed here is essentially independent of angular resolution .
based on angular resolution alone with a 2sr field - of - view hawc ( and milagro ) remove roughly 99.93% of the background in their field - of - view . the background rejection discussed below
is based purely on differences in the air showers regardless of their incoming direction .
figure [ fig : events ] shows three gamma - ray induced events and three proton induced events in hawc .
the figure shows the pulse height distribution of the pmts in the bottom layer of the detector .
the area of each square is proportional to the number of detected photoelectrons ( pes ) in the pmt .
the small black dots represent the pmts and the square is the position of the shower core .
the figure caption gives the primary energy for each event .
there are clear differences in morphology between proton and gamma - ray events .
though there are some proton induced events that have a similar morphology to gamma - rays induced events , the events shown here are typical in the sense that most events have similar properties .
note that while some of these events have very low primary energies , they are easily detected in hawc , striking many pmts .
-0.3 in , title="fig : " ] milagro utilizes a parameter known as compactness that is defined as the ratio of the number of pmts in the bottom layer with more than 2 pes detected to the maximum number of pes measured in a single pmt in the bottom layer@xcite .
hadronic showers tend to have small values of compactness and gamma ray showers a large value of compactness .
the studies performed here indicate that as one moves to higher elevations the ability of this parameter to discriminate between hadronic and gamma - ray initiated cascades degrades .
this is consistent with the observations of the milagro collaboration , who found that the background rejection capabilities of compactness improved as the zenith angle of the primary particle increased@xcite .
however , a modified form for the compactness parameter , where the number of pmts with more than 2 pes is replaced with the number of pmts that are struck in the bottom layer , has a similar background rejection capability to that obtained by milagro .
figure [ fig : hawc - compactness4572 ] shows this modified compactness parameter ( which will be referred to simply as compactness in what follows ) for proton ( dashed line ) and gamma ray ( solid line ) showers . if events are required to have a compactness greater than 8.3 one retains 66% of the gamma ray events and only 17% of the proton events , yielding a quality factor of 1.6 .
( the quality factor is defined as the fraction of gamma - ray events that survive the cut divided by the square root of the number of proton events that survive the cut , and is the relative improvement in sensitivity of the detector . )
another parameter that appears promising in discriminating gamma rays from background is the ratio of the number of pmts struck in the top layer ( @xmath12 ) to the number of pes in the brightest pmt in the bottom layer , where the pmts within 20 m of the fit shower core are excluded from the search for the maximum ( @xmath13 ) .
this distribution is shown in figure [ fig : hawc - nxtop ] for gamma ray and proton induced events .
the requirement @xmath14 retains 97.7% of the gamma ray events while rejecting 52.5% of the proton induced events .
the important feature of this parameter is that essentially all of the gamma ray events survive , so even though the quality factor is only 1.4 one has retained all of the signal events ( regardless of their energy ) .
the requirement @xmath15 yields a quality factor of 1.6 ( retaining 85% of the gamma ray induced events and 29% of the proton induced events ) , similar to the compactness criterion , but retaining a larger fraction of the gamma ray events .
unlike the compactness parameter defined above ( and that used by milagro ) there is very little energy dependence to this criterion . in milagro compactness
rejects low - energy gamma rays@xcite .
-0.3 in , title="fig : " ] -0.3 in for proton induced events and gamma - ray induced events .
see the text for the definitions of the parameters .
[ fig : hawc - nxtop],title="fig : " ] the performance of any detector is determined by its effective area as a function of energy , its angular resolution , the level of the background and the ability to eliminate this background .
these considerations can be parameterized in the following function : @xmath16 where @xmath17 is the `` sensitivity '' of the detector , @xmath18 is the angular resolution of the detector as a function of energy and zenith angle , @xmath19 is the differential spectral index of the source , @xmath20 is the effective area of the detector to gamma rays as a function of energy and zenith angle , @xmath21 is the zenith angle of the source as a function of time , @xmath22 is the effective area of the detector to the cosmic - ray background as a function of energy and zenith angle , and @xmath23 is the observation time . in determining the functions @xmath24 the background rejection should be included .
the time dependence of the zenith angle is required since the energy response of the detector is dependent upon the atmospheric overburden . to obtain the sensitivity of hawc , equation [ eqn : sensitivity ]
is evaluated with the aid of the monte carlo .
events are generated on an e@xmath11 spectrum for gamma ray primaries and @xmath25 for proton primaries .
after following the particles and the cherenkov light through the detector a list of pmts that are struck , the number of pes in each of these pmts , and the arrival time of the first pe in each pmt is written to a file .
the events are generated over a distribution of zenith angles to mimic an isotropic flux .
for a given source declination the time spent at each zenith angle is found and the effective area as a function of energy for that zenith angle is integrated to give a number of detected events as a function zenith angle for a given source declination .
figure [ fig : hawc - effective_area ] shows the effective area as a function of energy for gamma - ray primaries for hawc at two different detector altitudes , 4572 m asl and 5200 m asl .
the zenith angle averaging in this figure is as if the source was spread uniformly over the sky between zenith angles of 0 and 45 degrees .
this figure is used for illustration only . to calculate the sensitivity to a source the integral in equation [ eqn : sensitivity ]
is evaluated over the source transit ( given the declination of the source ) . in the remainder of the paper ( with the exception of the calculation of the sensitivity to gamma - ray bursts )
it is assumed that the source is at the declination of the crab ( 22 degrees ) and the detector is at a latitude of 36 degrees north .
only events that are reconstructed within 1.2 degrees of their true direction are included in the calculation of the effective area .
( this accounts for the first term @xmath18 in equation [ eqn : sensitivity ] . ) at high energies ( @xmath01 tev ) hawc at 5200 m would have about 170 times the effective area of milagro to gamma rays and at 100 gev about 1000 times the effective area of milagro .
as the energy decreases below this , the ratio of effective areas continues to increase , though there are insufficient monte carlo events at low energies for milagro to determine the ratio below 100 gev .
-0.3 in 50 pmts struck in the top layer , @xmath26 pmts used in the angular reconstruction , and the reconstructed angle within 1.2 degrees of the true direction.[fig : hawc - effective_area],title="fig : " ] not withstanding the above discussion , the `` sensitivity '' of a detector is a poorly defined concept .
since different astrophysical sources have different properties ( energy spectrum , duration , prior knowledge of their existence ) a single number called sensitivity is not useful . in
what follows the sensitivity of hawc to three distinct phenomena is estimated : galactic sources with an energy spectrum of @xmath27 , active galaxies at various redshifts ( to investigate the effect of the absorption of vhe photons by the extragalactic background light ) , and gamma ray bursts .
the latter is a major motivation for this type of detector , since acts by their nature have no opportunity to the prompt phase of grb emission .
the crab nebula is the prototypical galactic tev gamma ray source . in reality
the spectrum of the crab nebula is not well represented by a power law over the energy range 10 gev to 10 tev . however , for simplicity and to enable a comparison to other instruments the spectrum of the crab nebula@xcite is assumed to be @xmath28 m@xmath29s@xmath30 over the energy range of hawc . using the procedure outlined above the number of detected gamma rays per source transit in hawc
is calculated . to estimate the background level in a bin of the same size with the same transit ,
the identical procedure is carried out with proton primaries ( the criterion that the events must be reconstructed with 1.2 degrees of their true direction is dropped ) .
the identical procedure is than performed for the milagro monte carlo .
the ratio of the simulated events observed in hawc to the simulated events observed in milagro is used to scale the actual number of events detected in milagro and predict the background rate in hawc . for an observation altitude of 5200 m the background rate will be @xmath085 times that of milagro and for an observation altitude of 4572 m the background rate will be @xmath060 times larger than milagro s .
these numbers apply to the raw data , before the application of any background rejection .
table [ tab : crab_rates ] gives the expected number of source and background events for a single transit of the crab .
the predicted excess for milagro is somewhat larger than is actually observed by milagro . with two years of data milagro observes roughly 4@xmath31 on the crab nebula .
the discrepancy is accounted for by three detector related effects ( the dead - time in milagro is about 12% , the on time is about 90% , and at any given time about 6% of the pmts are not working ) and one astrophysical effect , the spectrum of the crab nebula turns over at the higher energies@xcite .
when these effects are properly accounted for the measurement of the crab flux by milagro is in good agreement with the measurement by hegra@xcite .
table [ tab : crab_rates ] shows that a detector at 5200 m elevation is about 40% more sensitive than the same detector at 4572 m elevation .
this is in agreement with the results given above on the effect of altitude on the particles at the ground . but both elevations result in a detector that can observe the crab nebula at high significance in a single transit .
this is an important feature which allows one to quickly verify all aspects of the detector response .
current acts such as the whipple telescope obtain roughly 5@xmath32 on the crab nebula in one hour of observation .
though the calculated transit here consists of 6 hours , the bulk of the signal arrives in a 4-hour span , so the sensitivity for the same observation period of hawc will be roughly 1/2 that of the whipple telescope . with one year of observation a 50 mcrab source would yield a 5@xmath32 detection .
in contrast the veritas 7-telescope array will detect a 7 mcrab source at 5@xmath32 with a 50 hour observation@xcite . with this sensitivity level veritas
would require 3 - 4 years of dedicated time to survey 2sr of the sky at the level of 50 mcrab .
( this assumes no change in sensitivity for veritas for sources 1 degree off - axis . ) it is worth noting that these two surveys are fundamentally different , in that hawc would obtain a measurement ( or upper limit ) which is an average over a year , while the veritas measurement ( or upper limit ) would result from a 7-minute snapshot over a 3 - 4 year period . given the transient nature of many vhe sources ,
such a snapshot may be a limited value . the gamma - ray rate ( in hawc ) from the crab will be roughly 0.4 hz or 24/minute for a detector at 5200 m and 0.23hz ( or 14/minute ) for a detector at 4572 m asl .
this number compares favorably with veritas and is significantly higher than that currently observed by the whipple telescope .
.expected signals from the crab nebula in hawc at 5200 m and 4572 m elevation .
milagro is at an elevation of 2700 m and is @xmath33 the size of hawc .
[ cols="^ , < , < , < , < , < , < " , ] [ tab : crab_rates ] for more distant sources one must account for the absorption of high - energy photons caused by interactions with the extragalactic background light ( ebl)@xcite . while a direct measurement of the ebl has proved elusive , several models exist . in
what follows the `` fast evolution '' model of stecker and de jager@xcite is used to estimate the effect as a function the redshift of the source on the sensitivity of hawc . for nearby objects
( redshift less than 0.05 ) there is a negligible difference between this model and the `` baseline '' model discussed in the reference . at a redshift of 0.2 the baseline model for the ebl results in @xmath015% more signal events detected by hawc than the model used here .
figure [ fig : hawc - ebl - energy ] shows the number of signal events detected as a function of energy for sources at various redshifts .
only events that satisfied the background rejection criterion , @xmath34 , had more the 20 pmts survive the angular reconstruction , and were reconstructed within 1.2 degrees of their true direction where used to calculate the curves in the figure . a source differential spectral index of -2.49 and an observation altitude of 5200 m
have been assumed in this figure .
an integration of these curves gives the total number of detected events for each source .
-0.3 in are used to calculate the curves in the figure.[fig : hawc - ebl - energy],title="fig : " ] using the modified spectra given in figure [ fig : hawc - ebl - energy ] the sensitivity of hawc to distant sources can be calculated .
figure [ fig : hawc - agn - time5 ] shows the number of days required to detect a source at the 5@xmath32 level as a function of source intensity for several different source redshifts .
the source intensity is given in units of the crab flux and is the flux that would be present at the top of the atmosphere if there was no absorption of the high - energy gamma rays .
hawc could detect a source at a redshift of 0.3 with a flux at the top of the atmosphere ( before accounting for absorption ) one tenth that of the crab nebula with less than one year of observation .
-0.3 in as a function of the source intensity for sources at various redshifts .
[ fig : hawc - agn - time5],title="fig : " ] the study of the high - energy spectrum of gamma - ray bursts ( grbs ) is perhaps the strongest motivation for an all - sky vhe telescope . while glast will have good sensitivity to grbs up to several 10 s of gev , a ground - based instrument , with its much larger effective area ,
could have good sensitivity up to the highest energy gamma rays that reach the earth . the ability to study the prompt phase of vhe emission could lead to exciting new physics .
given its low energy threshold , hawc would have sensitivity to grbs at a redshift 1 . to estimate the sensitivity of hawc to grbs requires knowledge of any absorption at the source and the inevitable absorption of the vhe photons as they traverse the cosmos . since one does not know the inherent spectrum of grbs , in what follows the grb energy spectrum
is modeled as a power law with a sharp cutoff .
this cutoff could arise from absorption at the source or from the ebl . by investigating the sensitivity of hawc to different cutoff energies it is possible to estimate the sensitivity to grbs at different redshifts .
figure [ fig : hawc_grb ] shows the sensitivity of hawc ( at an observation altitude of 4572 m asl ) to grbs as a function of burst duration for several energy cutoffs .
the spectra are assumed to be @xmath3 up to the indicated cutoff energy , where the photon flux falls to zero .
these curves give the flux level needed to observe a grb at the 5@xmath32 level .
no background rejection has been applied to these data .
the black circles are data from the batse detector , giving the observed distribution of grb fluence as a function of burst duration .
the lower solid line is the sensitivity of glast and the upper solid line the sensitivity of egret . to detect a 1-second duration burst hawc
would require 50 events from a grb over a background level of 85 events .
the signal to noise level in a detected burst would be quite high .
-0.3 in detection in hawc for various cutoff energies ( as given in the legend ) . for comparison
the sensitivities of egret and glast are shown.[fig : hawc_grb],title="fig : " ] figure [ fig : hawc_grb ] can be used in conjunction with the known redshift distribution of grbs and the known grb rate to estimate how many grbs / year hawc would detect . assuming that the intrinsic grb spectrum extends to at least 50 gev
, one sees that roughly 20% of all grb s are bright enough to be detected by hawc .
if half of all bursts lie within a redshift of 1 and assuming a grb rate of 1 per day , hawc ( which views 1/6 of the sky ) would detect at least 6 grbs per year .
if the intrinsic grb spectra extend beyond 50 gev hawc would detect more bursts .
a design for the next generation all - sky vhe gamma - ray telescope has been presented .
this instrument , hawc , would be over 20 times more sensitive than the milagro detector . with the ability to continuously view the entire overhead sky hawc will be an excellent complement to both glast and the coming generation of air cherenkov telescopes ( hess , magic , veritas , and cangaroo iii ) . with comparable sensitivity to glast it will be the only instrument capable of monitoring the many thousands of sources that glast is expected to detect at higher energies .
in addition to searching the sky for galactic sources ( the vhe complement to the 150 egret unidentified objects ) , and active galaxies , the low - energy sensitivity of a detector placed at high altitude ensures that such an instrument will detect any vhe emission from gamma - ray bursts .
perhaps most importantly an open aperture instrument with this level of sensitivity could discover completely new and unexpected phenomena that have so far eluded detection . 0 aharonian , f.a .
, et al . , 2000 , _ apj _ , * 539 * , 317 .
alexandreas , d.e . , et al . , 1993 , _
nim _ , * a328 * , 570 .
amenomori , m , et al . , 1999 , _ apj _ , * 525 * , l93 .
atkins , r.a .
, et al . , 2000 , _ apj _ , * 533 * , l119 .
atkins , et al .
, 2003 , _ apj _ , * 595 * , 803 .
gaidos , j. , et al . , 1996 , _
nature _ , * 383 * , 319 .
hanna , d.s .
, et al . , 2002 , _
nim _ , * a491 * , 126 .
hartman , r.c .
, et al . , 1999 , _
apjs _ , * 123 * , 79 .
hillas , a.m. , et al .
, _ apj _ , * 503 * , 744 .
hofmann , w. , et al .
, 2003 , _ proceedings of the 28@xmath35 icrc _ , * 5 * , 2811 .
horan , d. and weekes , t.c .
, 2003 , astro - ph/030391v1 .
knapp , j. and heck , a. , 1993 , extensive air shower simulation with corsika : a user s manual ( kfk 5196 b ; karlsruhe : kernforschungszentrum karlsruhe ) .
martnez , m. , et al . , 2003 , _ proceedings of the 28@xmath35 icrc _ , * 5 * , 2815 .
par , e. , et al . , 2002 ,
_ nim _ , * a490 * , 71 .
primack , j. r. , somerville , r. s. , bullock , j. s. , and devriendt , j. e. g. , 2000 , aip conference proceedings , 558 , 463 , aip : f. a. aharonian and h. j. v " olk stecker , f. and de jager , o. c. , 2002 , _ apj _ , * 566 * , 738 .
weekes , t.c .
, et al . , 1989 , _
apj _ , * 342 * , 379 .
weekes , t.c .
, et al . , 2002 , _ astroparticle physics _ , * 17 * , 221 .
yoshikoski , t , et al . , 1999 , _
astroparticle physics _ , * 11 * , 267 . | the study of the universe at energies above 100 gev is a relatively new and exciting field .
the current generation of pointed instruments have detected tev gamma rays from at least 10 sources and the next generation of detectors promises a large increase in sensitivity .
we have also seen the development of a new type of all - sky monitor in this energy regime based on water cherenkov technology ( milagro ) . to fully understand the universe at these extreme energies requires a highly sensitive detector capable of continuously monitoring the entire overhead sky .
such an instrument could observe prompt emission from gamma - ray bursts and probe the limits of lorentz invariance at high energies .
with sufficient sensitivity it could detect short transients ( @xmath015 minutes ) from active galaxies and study the time structure of flares at energies unattainable to space - based instruments . unlike pointed instruments a wide - field instrument can make an unbiased study of all active galaxies and enable many multi - wavelength campaigns to study these objects .
this paper describes the design and performance of a next generation water cherenkov detector . to attain a low energy threshold and have high sensitivity
the detector should be located at high altitude ( @xmath1 4 km ) and have a large area ( @xmath040,000 m@xmath2 ) .
such an instrument could detect gamma ray bursts out to a redshift of 1 , observe flares from active galaxies as short as 15 minutes in duration , and survey the overhead sky at a level of 50 mcrab in one year . |
the taub - nut ( newman , unti and tamburino ) space - time @xcite is a stationary , spherically symmetric and non - asymptotically flat solution of the vacuum einstein equation in general theory of relativity .
the space - time has topology @xmath0 with lorentzian signature @xcite .
the nut space - time is described by two parameters : one is the mass parameter @xmath1 and another one is the nut parameter @xmath2 .
there is no modification required in the einstein - hilbert action to accommodate the nut charge @xcite or `` dual mass '' @xcite or `` gravito - magnetic mass '' or `` gravito - magnetic monopole '' @xcite .
this dual mass is an intrinsic feature of general theory of relativity .
the space - time contains closed time - like curve and null lines .
it is a geodetically incomplete space - time @xcite .
bonor @xcite has given a new interpretation of the nut spacetime and it describes ` the field of a spherically symmetric mass together with a semi - infinite massless source of angular momentum along the axis of symmetry ' . on the other hand , manko and
ruiz @xcite analyzed the mass and angular momentum distributions in case of generalized nut spacetime using komar integral approach .
t hooft and polykov @xcite have demonstrated that the magnetic monopole present in certain non - abelian gauge theories .
zee @xcite observed that there is an existence of a gravitational analog of dirac s magnetic monopole @xcite .
the author is also discussed regarding the mass quantization following the idea of dirac quantization rule .
he also claimed that there is certainly no experimental evidence of mass quantization .
moreover , he proposed that if mass is quantized there may have profound consequences in physics .
for example , if a magnetic monopole moving around a nucleus then the mass quantization rule suggests that the binding energy of every level in the nucleus is also quantized .
friedman and sorkin @xcite observed that the gravito - pole may exist in topological solution .
dowker @xcite proposed that the nut spacetime as a ` gravitational dyon ' . the euclidean version of the space - time is closely related to the dynamics of bps ( bogomolnyi - prasad - sommerfield ) monopoles @xcite . the experimental evidence of this dual mass has not been verified till now .
there may be a possibilities of experimental evidences in near future and it was first proposed by lynden - bell and nouri - zonoz @xcite in 1998 .
letelier and vieira @xcite have observed that the manifestation of chaos for test particles moving in a tn space - time perturbed by dipolar halo using poincare sections . the geodesics structure in euclidean
tn space - time has been studied in the ref .
@xcite .
the gravito - magnetic lensing effect in nut space - time was first studied by nouri - zonoz et al .
@xcite in 1997 .
they proved that all the geodesics in nut spacetime confined to a cone with the opening angle @xmath3 defined by @xmath4 where @xmath5 is the impact factor . ) for null circular geodesics .
] for small @xmath6 and in the limit @xmath7 , it should be @xmath8 it should also be noted that the opening angle is proportioal to the nut parameter @xmath2 .
furthermore , they also examined the lensing of light rays passing through the nut deflector .
this properties modified the observed shape , size and orientation of a source .
it has been also studied there that there is an extra shear due to the presence of the gravito - magnetic monopole , which changes the shape of the source . the same author also studied the electromagnetic waves in nut space through the solutions of the maxwell equations via newman - penrose null tetrad formalism to further deeper insight of the physical aspects of the dual mass .
since the tn space - time has gravito - magnetic monopole that changes the structure of the accretion disk and might offer novel observational prospects @xcite . the maximal analytic extension or kruskal
like extension of the tn space - time shows that it has some unusual properties @xcite .
maximal analytic extension is needed in order to understand the global properties of the space - time .
misner and taub have shown that tn space is maximally analytic i.e. it has no hausdorff extension @xcite . whereas hajicek @xcite showed that the non - hausdorff property occurs only on the killing horizons and causes no geodesics to bifurcate .
chakraborty and majumdar @xcite have derived the exact lense - thirrring precession ( inertial frame dragging effect ) in case of the tn space - time in comparison with the mass - less tn space - time . the _ mass - less dual mass _
( i.e. tn space - time with @xmath9 ) concept was first introduced by ramaswamy and sen @xcite .
they also proved that ` in the absence of gravitational radiation magnetic mass requires either that the metric be singular on a two dimensional world sheet or the space - time contain closed time - like lines , violating causality ' .
after that ashtekar and sen @xcite demonstrated that the consequences of magnetic mass in quantum gravity .
they also proved that the dual mass implies the existence of ` wire singularities ' in certain potentials for weyl curvature .
finally , mueller and perry @xcite showed that the ` mass quantization ' rule regarding the nut space - time . in @xcite
, the author has been studied @xmath10 time - dependent tensorial perturbations of lorentzian tn space - time and proved that lorentzian tn space - time is unstable .
geodesics of accelerated circular orbits on the equatorial plane has been studied in detail of the nut space using frenet - serret procedure @xcite .
however , in the present work we wish to investigate the complete geodesic structure of the tn space - time in the equatorial plane .
we compare the circular geodesics in the tn space - time with mass - less tn space - time and zero nut parameter by _ analyzing the effective potential graphically for both null cases and time - like cases_. the presence of the dual mass can changes the geodesic structure in comparison with the mass - less dual mass and zero dual mass .
this is clearly manifested in the effective potential diagram .
we also differentiate _ graphically _ the isco , mbco and cpo of the said space - time in comparison with the mass - less cases .
moreover , we examine the circular geodesics in the @xmath11 plane for different values of energy i.e. @xmath12 , @xmath13 and @xmath14 in case of tn and mass - less tn spacetime , and plotted graphically .
furthermore , we have studied more exotic cases i.e. extreme cases , where we find surprising results that the radii of three important class of orbits namely , the radius of isco ( @xmath15 ) , the radius of mbco ( @xmath16 ) and the the radius of cpo ( @xmath17 ) are coincident with the horizon . from the best of my knowledge ,
this is the first we have observed in this work for any spherically symmetric , stationary and _ non - asymptotically _ flat spacetime showing such feature . finally , we compute the cm energy for these spacetimes and we have found that for non - extreme tn bh , the cm energy is finite whereas for _ extreme _ tn bh the cm energy is diverging in nature .
the circular geodesics studied earlier by several authors @xcite but they have not been studied in more graphically . we here show the differences between the two spacetime with the mass parameter and the mass - less parameter in visually .
it may be noted that circular orbits of arbitrary radii are not possible , there exists a minimum radius below which no circular orbits are possible .
the geodesic structure has been studied earlier for schwarzschild bh @xcite , reissner nordstrm ( rn ) bh @xcite , kerr - taub - nut ( ktn ) bh @xcite and kerr - newman - taub - nut ( kntn ) more recently @xcite . in this work ,
we have specialized on the cases when the parameter @xmath18 . by studying the geodesic structure
, we can extract more information about the back ground space - time .
different observables like lense - thirrring effect , gravitational time delay , gravitational bending of light etc .
all are the phenomenon related to the geodesic structure of the space - time .
this is one of the major motivation behind to study them .
the structure of the manuscript is as follows . in section [ nut ]
, we discuss the basics of tn bh . in section [ egtn ] ,
we study the equatorial geodesic properties of the said bh .
section [ ex ] devoted to study the extreme tn bh .
the conclusions are given in the section [ dis ] .
the metric is given by @xcite @xmath19 where , @xmath1 denotes the gravito - electric mass or adm mass and @xmath2 denotes the gravito - magnetic mass or dual mass or magnetic mass of the space - time .
it is clearly evident that there are two types of singularities are present in the metric ( [ tn1 ] ) .
one is at @xmath20 which give us the killing horizons or bh horizons : @xmath21 @xmath22 is called event horizon and @xmath23 is called cauchy horizon . from fig .
[ eq1 ] , we can see the horizon structure of tn and mass - less tn bh .
there is a qualitative difference between two horizon structure with nut parameter and without nut parameter .
, title="fig:",scaledwidth=45.0% ] , title="fig:",scaledwidth=45.0% ] in the limit @xmath24 , one obtains the schwarzschild bh .
interestingly , when @xmath9 , we get mass - less tn space - time and the horizons at @xmath25 the other type of singularity occurs at @xmath26 and @xmath27 , where the determinant of the metric component vanishes .
misner@xcite first demonstrated that in order to remove the apparent singularities at @xmath26 and @xmath27 , @xmath28 must be identified modulo @xmath29 .
provided that @xmath30 .
it should be noticed here that the nut parameter actually measures deviation from the asymptotic flatness at infinity which may be manifested in the off - diagonal components of the metric and this is happening due to presence of the dirac - misner type of singularity .
when , @xmath31 the tn metric describes a bh , otherwise it has a naked singularity .
when @xmath32 , we find extreme tn bh .
the characteristics of the variation of @xmath33 with radial coordinate is shown in fig .
[ eq2 ] .
with @xmath34 for tn and massless
tn bh . [ eq2],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn and massless
tn bh . [ eq2],title="fig:",scaledwidth=45.0% ] the `` red - shift factor '' ( @xmath35 ) @xcite for tn bh is given by @xmath36 and for mass - less tn bh , it is @xmath37 the variation of redshift factor with radial coordinate is shown in fig .
[ eqrs ] . for tn and massless
tn bh . [ eqrs],title="fig:",scaledwidth=45.0% ] for tn and massless
tn bh . [ eqrs],title="fig:",scaledwidth=45.0% ] the `` red - shift''(@xmath38 ) @xcite is given by @xmath39 for mass - less tn bh , it is @xmath40 the variation of redshift with radial coordinate is depicted in fig .
[ eqz ] . for tn and massless
tn bh . [ eqz],title="fig:",scaledwidth=45.0% ] for tn and massless
tn bh . [ eqz],title="fig:",scaledwidth=45.0% ] the thermodynamic properties of tn bh could be found in detail in @xcite and the bh temperature of @xmath41 denotes event horizon of the bh .
] was calculated there @xmath42 in the limit @xmath43 , @xmath44 , this indicates that there must exists _ extreme _ tn bh .
the geodesic properties have been studied in sec .
since the equatorial plane is the good location where we can see the causal characteristics of the geodesics so in this section we would like to study them . it should be noted that @xmath45 , a constant is called circular geodesics .
also , the study of geodesics in the tn space - time is governed by the laws of conservation of energy and angular momentum because the space - time is independent of the coordinates @xmath28 and @xmath46 .
the tn space - time possesses time - like isometry generated by the time - like killing vector @xmath47 whose projection along the four velocity @xmath48 of geodesics : @xmath49 , is conserved along such geodesics and the another conserved quantity is angular momentum followed by the relation @xmath50 ( where @xmath51 ) . where @xmath52 is the space - like killing vector field due to the rotational isometry . using these conditions together with the normalization of the four velocity
, one can easily derived the radial equation for tn bh on the @xmath53 plane : @xmath54 where the effective potential @xcite is @xmath55 here , @xmath56 for time - like geodesics , @xmath57 for light - like geodesics and @xmath58 for space - like geodesics . for light rays orbit ,
the effective potential becomes @xmath59 first we see the behaviour of the test particle in the potential well diagram . in fig .
[ nu1 ] , fig .
[ nu2 ] , fig .
[ nu3 ] , fig .
[ nu4 ] , fig .
[ nu5 ] , fig .
[ nu6 ] , fig .
[ nu7 ] and fig .
[ nu8 ] , we show how the effective potential for photon changes for different values of angular momentum parameter and nut parameter . from the effective potential diagram , it has been observed that the presence of _ dual mass _ parameter effectively changes the shape of the potential well in comparison with absence of the _ dual mass _ parameter .
the structure of the potential well also changes in the presence of adm mass parameter and in the absence of adm mass parameter .
with @xmath34 for tn bh and mass - less tn bh .
[ nu1],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn bh and mass - less tn bh .
[ nu1],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn bh and mass - less tn bh for different values of @xmath2 .
[ nu2],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn bh and mass - less tn bh for different values of @xmath2 .
[ nu2],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn bh and mass - less tn bh .
[ nu3],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn bh and mass - less tn bh .
[ nu3],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn bh and mass - less tn bh .
[ nu4],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn bh and mass - less tn bh .
[ nu4],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn bh and mass - less tn bh .
[ nu5],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn bh and mass - less tn bh .
[ nu5],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn bh and mass - less tn bh .
[ nu6],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn bh and mass - less tn bh .
[ nu6],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn bh and mass - less tn bh [ nu7],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn bh and mass - less tn bh [ nu7],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn bh and mass - less tn bh .
[ nu8],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn bh and mass - less tn bh .
[ nu8],title="fig:",scaledwidth=45.0% ] now by introducing the impact parameter @xmath5 , one can reparametrization of any null geodesics described by the parameter @xmath60 and @xmath61 .
@xmath62 be the angular momentum of null geodesics when it is reparametrized to have unit energy .
now the plot of @xmath62 with @xmath34 gives the information of radial motion for null geodesics .
the equations evaluating the radius @xmath63 of the unstable circular photon orbit at @xmath64 and @xmath65 by introducing the impact parameter @xmath66 are @xmath67d_{c}^2 & = & 0 ~.\label{n4}\end{aligned}\ ] ] from eq .
( [ n4 ] ) , we find @xmath68 the behaviour of the impact parameter can be shown from the fig .
[ nu10 ] .
with @xmath63 for tn bh and mass - less tn bh .
[ nu10],title="fig:",scaledwidth=45.0% ] with @xmath63 for tn bh and mass - less tn bh .
[ nu10],title="fig:",scaledwidth=45.0% ] putting the eq .
( [ n5 ] ) in eq .
( [ n3 ] ) , we obtain the equation of cpo @xcite : @xmath69 for mass - less tn bh @xcite , the root of the eq .
becomes @xmath70 the variation of cpo with @xmath63 for tn bh and mass - less tn bh could be found in the fig .
[ nu10 ] . .
for mass - less cases , @xmath71 . for @xmath24
this root has the value @xmath72 .
[ nu11],title="fig:",scaledwidth=45.0% ] . for mass - less cases , @xmath71 . for @xmath24
this root has the value @xmath72 .
[ nu11],title="fig:",scaledwidth=45.0% ] the gravitational bending of light analyzed in @xcite and the bending angle on the cone is computed in @xcite @xmath73 . in terms of opening angle
it should be @xmath74 . for _ massless tn spacetime
, this angle reduces to zero value_. for time - like geodesic we have to set @xmath56 then the effective potential becomes @xmath75 the qualitative behaviour of the test particle can be obtain by studying this potential .
first we consider the zero angular momentum geodesics for this the effective potential reduces to @xmath76 one can observe the qualitative behaviour of the geodesics in the presence of dual mass and with out dual mass and also the mass - lees parameter .
this can be seen from the fig .
[ t1 ] .
with @xmath34 for tn bh and mass - less tn bh .
[ t1],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn bh and mass - less tn bh .
[ t1],title="fig:",scaledwidth=45.0% ] in this plot , we can see that the presence of the dual mass parameter deforms the shape of the radial effective potential in comparison with zero dual mass parameter .
this behaviour can be seen from the mass - less case also .
for @xmath77 , the behaviour of the test particle in the potential well could be seen from the following fig .
[ t2 ] , fig .
[ t3 ] , fig .
[ t4 ] , fig .
[ t5 ] , fig .
[ t6 ] , fig .
[ t7 ] , fig .
[ t8 ] and fig .
[ t9 ] .
with @xmath34 for tn bh and mass - less tn bh .
[ t2],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn bh and mass - less tn bh .
[ t2],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn bh and mass - less tn bh .
nut parameter .
[ t3],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn bh and mass - less tn bh .
nut parameter .
[ t3],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn bh and mass - less tn bh .
[ t4],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn bh and mass - less tn bh .
[ t4],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn bh and mass - less tn bh .
[ t5],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn bh and mass - less tn bh .
[ t5],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn bh and mass - less tn bh .
[ t6],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn bh and mass - less tn bh .
[ t6],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn bh and mass - less tn bh .
[ t7],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn bh and mass - less tn bh .
[ t7],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn bh and mass - less tn bh .
[ t8],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn bh and mass - less tn bh .
[ t8],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn bh and mass - less tn bh .
[ t9],title="fig:",scaledwidth=45.0% ] with @xmath34 for tn bh and mass - less tn bh .
[ t9],title="fig:",scaledwidth=45.0% ] in the above figures , we have seen that the radial dependency of the effective potential with dual mass parameter , without dual mass parameter and @xmath9 .
when @xmath24 , the effective potential at large radial distance does not change much more with the increasing of @xmath60 . when we introduced the nut parameter , the shape of the effective potential _
deforms _ in comparison with nut less case and it also changes for different values of angular momentum parameter . finally , when we increase the value of @xmath2 the height of the potential well decreases .
this work has been done earlier for the parameter @xmath78 , @xmath79 and @xmath2 @xcite .
this work is specially for the parameter values @xmath18 and for mass - less cases which has been not studied previously .
now re - write the eq .
( [ tnr ] ) for time - like geodesics as @xmath80 @xmath81 for circular geodesics @xmath45 , we have the following condition : @xmath82 from this condition , we find the energy and angular momentum for circular orbit : @xmath83 and @xmath84 these equations require for energy square and angular momentum square positive definite i.e. @xmath85 . compared with the eq .
( [ n6 ] ) , it implies that the minimum radius for time - like circular orbit is the radius of of the unstable cpo .
interestingly for mass - less case , these values are @xmath86 and @xmath87 we have plotted their behaviour in the fig .
[ ff ] and fig .
[ ff1 ] . with @xmath88 for tn bh and mass - less tn bh .
[ ff],title="fig:",scaledwidth=45.0% ] with @xmath88 for tn bh and mass - less tn bh .
[ ff],title="fig:",scaledwidth=45.0% ] and with @xmath88 for tn bh and mass - less tn bh .
[ ff1],title="fig:",scaledwidth=45.0% ] with @xmath88 for tn bh and mass - less tn bh .
[ ff1],title="fig:",scaledwidth=45.0% ] from eq .
( [ n10 ] ) , and for circular orbit one obtains @xmath89 it should be noted that @xmath90 . from this expression , it follows that the angular momentum parameter explicitly dependes upon the energy value .
therefore there must be difference in the circular orbits in the @xmath11 plane for @xmath12 , @xmath13 and @xmath14 .
we are interested here to look the behaviour of the circular geodesics in this plane by incorporating these energy conditions . when @xmath13 , we obtain the bound orbits , this can displayed from the @xmath91 diagram .
it has been shown in the fig .
[ t10 ] .
with @xmath88 for tn bh and mass - less tn bh .
[ t10],title="fig:",scaledwidth=45.0% ] with @xmath88 for tn bh and mass - less tn bh .
[ t10],title="fig:",scaledwidth=45.0% ] for @xmath14 , we find the marginally escape orbits .
this has been seen from the fig .
[ t11 ] .
with @xmath88 for tn bh and mass - less tn bh .
[ t11],title="fig:",scaledwidth=45.0% ] with @xmath88 for tn bh and mass - less tn bh .
[ t11],title="fig:",scaledwidth=45.0% ] finally for @xmath12 , we find the unbound orbits .
this could be seen from the fig .
[ t12 ] , fig .
[ t13 ] and fig .
[ t14 ] .
with @xmath88 for tn bh and mass - less tn bh .
[ t12],title="fig:",scaledwidth=45.0% ] with @xmath88 for tn bh and mass - less tn bh .
[ t12],title="fig:",scaledwidth=45.0% ] with @xmath88 for tn bh and mass - less tn bh .
[ t13],title="fig:",scaledwidth=45.0% ] with @xmath88 for tn bh and mass - less tn bh .
[ t13],title="fig:",scaledwidth=45.0% ] with @xmath88 for tn bh and mass - less tn bh .
[ t14],title="fig:",scaledwidth=45.0% ] with @xmath88 for tn bh and mass - less tn bh .
[ t14],title="fig:",scaledwidth=45.0% ] from fig .
[ t15 ] , fig .
[ t16 ] , fig .
[ t17 ] , fig .
[ t18 ] and fig .
[ t19 ] , one can observe that the variation of @xmath92 with respect to @xmath88 for various values of nut parameter in the presence of mass parameter and without mass parameter .
now we see the behaviour of the @xmath60 with @xmath34 for different values of energy for a fixed value of the dual mass parameter .
with @xmath88 for tn bh and mass - less tn bh .
[ t15],title="fig:",scaledwidth=45.0% ] with @xmath88 for tn bh and mass - less tn bh .
[ t15],title="fig:",scaledwidth=45.0% ] with @xmath88 for tn bh and mass - less tn bh .
[ t16],title="fig:",scaledwidth=45.0% ] with @xmath88 for tn bh and mass - less tn bh .
[ t16],title="fig:",scaledwidth=45.0% ] with @xmath88 for tn bh and mass - less tn bh .
[ t17],title="fig:",scaledwidth=45.0% ] with @xmath88 for tn bh and mass - less tn bh .
[ t17],title="fig:",scaledwidth=45.0% ] with @xmath88 for tn bh and mass - less tn bh .
[ t18],title="fig:",scaledwidth=45.0% ] with @xmath88 for tn bh and mass - less tn bh .
[ t18],title="fig:",scaledwidth=45.0% ] with @xmath88 for tn bh and mass - less tn bh .
[ t19],title="fig:",scaledwidth=45.0% ] with @xmath88 for tn bh and mass - less tn bh .
[ t19],title="fig:",scaledwidth=45.0% ] now the isco radius can be obtain by using the condition eq .
( [ ef ] ) with additional condition : @xmath93 although it has been derived earlier in @xcite .
we are interested here to see the behaviour of isco in the presence of nut parameter and with out nut parameter in comparison with mass - less cases in more graphically which has been not shown there . therefore the isco equation for tn bh
could be obtain by applying the condition given in eq .
( [ ef ] ) and eq .
( [ n15 ] ) ( or by putting @xmath18 in eq .
( 81 ) in @xcite ) : @xmath94 @xmath95 and for the mass - less tn bh , it is @xmath96 and it also indicates there is _ no isco for massless tn bh_. it is a curious result although there exists mass - less cpo .
the variation of isco in the presence of the nut parameter and without nut parameter could be found in the fig .
[ t20 ] . in the presence of nut parameter and without nut parameter . where @xmath97 . [ t20],title="fig:",scaledwidth=45.0% ] another interesting orbit that has not been considered previously in @xcite but considered in @xcite for kntn
so we have just set @xmath98 in eq .
( [ ef1 ] ) ( or putting the parameters @xmath18 in eq .
( 98 ) in @xcite ) , we find the mbco for tn bh : @xmath99 @xmath100 and for the mass - less tn bh , it is @xmath101 it implies that there is no existence of mbco for mass - less case . from eq .
( [ n18 ] ) , in the limit @xmath24 , we get the mbco for schwarzschild bh @xcite . in fig .
[ t21 ] , we have plotted the mbco with @xmath88 in comparison with schwarzschild bh . .
[ t21],title="fig:",scaledwidth=45.0% ]
in the previous section , we have discussed the complete geodesic structure for non - extreme tn bh . moreover , we have clearly explained the difference of geodesic structure between tn spacetime and mass - lees tn spacetime in graphically . in the present section
we shall discuss more interesting case i.e. _ extreme tn _ spacetime .
what happens the geodesic structure in the extreme limit i.e. @xmath43 .
this is the main aim in this section .
proceeding similarly , we can write the effective potential for massive particles in the extreme limit [ using eq .
( [ n8 ] ) ] given by @xmath102 the qualitative behaviour of the test particle may be seen from the effective potential diagram ( see fig .
[ t22 ] ) .
with @xmath34 for extreme tn bh , here @xmath103 .
[ t22],title="fig:",scaledwidth=45.0% ] proceeding analogously , we apply for circular geodesics @xmath45 , one obtains the energy and angular momentum for extreme tn bh [ using eq .
( [ ef1 ] ) and eq .
( [ ef2 ] ) ] : @xmath104 and @xmath105 how @xmath106 and @xmath107 are varied with @xmath88 , it can be seen from the fig .
[ t23 ] . for extreme tn bh , here @xmath103 .
[ t23],title="fig:",scaledwidth=45.0% ] for extreme tn bh , here @xmath103 .
[ t23],title="fig:",scaledwidth=45.0% ] now we have derived the isco equation for non - extreme tn bh but here we will see what would be the isco radius in the extremal limit which is most important class of circular orbit in astrophysics .
we find for extreme tn bh the isco radius [ using eq .
( [ n16 ] ) ] should be @xmath108 the corresponding isco energy and isco angular momentum are @xmath109 for photons , the effective potential in the extreme limit [ using eq .
( [ n1 ] ) ] reduces to @xmath110 its behaviour for different values of angular momentum parameter can be seen from the fig .
[ t24 ] .
with @xmath34 for extreme tn bh , here @xmath103 .
[ t24],title="fig:",scaledwidth=45.0% ] therefore , one obtains the cpo [ using eq .
( [ n6 ] ) ] for extreme tn bh : @xmath111 similarly , we can find the radius of mbco [ using eq .
( [ n18 ] ) ] for extreme tn bh : @xmath112 interestingly , we observed that for extreme tn bh the three radii namely isco , mbco and cpo coincides with the horizon i.e. @xmath113 this has not been previously examined in the literature that for extreme tn bh three orbits are coincident with the killing horizons i.e. null geodesic generators of the horizon . for
spherically symmetric extreme string bh , this result has been observed in @xcite .
probably , this is a _ first _ time we have seen that such type of features for any spherically symmetric , stationary and non - asymptotic flat extreme bh .
in this paper we examined the geodesic motion of test particles in the background geometry of tn spacetime in comparison with mass - less ( zero mass ) tn spacetime .
we considered the both massive and mass - less cases .
we differentiated the isco , mbco and cpo in graphically between tn spacetime and mass - less tn spacetime . from effective potential diagram
, we showed the presence of the nut parameter changes the shape of the potential well in comparison with zero nut paprameter and mass - less spacetime .
we studied the circular orbits in the @xmath11 plane for different values of energy i.e. @xmath12 , @xmath13 and @xmath14 in case of tn and mass - less tn spacetime .
we also derived the geodesic motion of test particles ( both massive and mass - less ) in case of extreme tn bh .
interestingly , we showed that the three radii i.e. the radius of isco , the radius of mbco and the radius of cpo are coincident with the killing horizon radius .
this is a very surprising result because probably we first obtained this result for any spherically symmetric , stationary and non - asymptotic flat spacetime . in appendix -
a , we derived the cm energy for extreme tn bh and we found that the diverging value of cm energy .
in this section , we should study what is the role of nut parameter in the baados , silk and west ( bsw ) effect which was predicted by baados , silk and west @xcite several years ago .
does tn bh could be act as a particle accelerator with arbitrarily high energy when the bh is extremal .
this is the main aim in this section and motivated by the previous section from the analysis of geodesic structure .
we have considered the particle acceleration and collision in the cm frame . to determine the cm energy , we consider first two particles coming from infinity with @xmath114 approaching the tn bh with different angular momenta @xmath115 and @xmath116 and colliding at some radius @xmath34 .
later , we choose the collision point is at @xmath34 to approach the horizon @xmath117 .
also we have assumed that the particles are initially to be at rest at infinity .
the formula that we have used here suggested first by bsw @xcite should read @xmath118 since in the previous section we have calculated total geodesic structure confined in the equatorial plane for tn bh , so we should not repeat it again .
due to the time - like isometry and space - like isometry the space - time possesses two conserved quantities one is the energy and other is the angular momentum .
thus for massive particles , the components of the four velocity are @xmath119 and putting these values in ( [ cme ] ) , we find the expression for cm energy : @xmath121 where , @xmath122 for simplicity , @xmath123 and reverting back the value of @xmath124 , we obtain the cm energy near the event horizon ( @xmath22 ) of tn space - time : @xmath125 and for cauchy horizon ( @xmath23 ) the cm energy is @xmath126 where @xmath127 is defined previously .
it implies that the cm energy is finite and depends upon the values of angular momentum parameter .
it also suggests that the cm energy depends upon the nut parameter .
it may also playing a key role in the bsw effect as we have seen from the above expression .
now see what happens in case of mass - less case and _ extreme _ case ?
first we consider the mass - less case , in this case the cm energy is given by @xmath129 it indicates that the cm energy is finite depends on the nut parameter .
more interesting case i.e. the extreme case where the cm energy is given by @xmath130 since we know for extreme case , @xmath131 and @xmath132 .
therefore we find for extreme tn bh : @xmath133 i.e the cm energy is _
divergent_. this is an amusing result . because we first reported a non - asymptotic flat , spherically symmetric and stationary extreme bh spacetime possesses such properties . | in this work we study the equatorial causal geodesics of the taub - nut ( tn ) space - time in comparison with _ mass - less _ tn space - time .
we emphasized both on the null circular geodesics and time - like circular geodesics . from the effective potential diagram of
null and time - like geodesics , we differentiate the geodesics structure between tn spacetime and mass - less tn space - time .
it has been shown that there is a key role of the nut parameter to changes the shape of pattern of the potential well in the nut spacetime in comparison with _ mass - less _ nut space - time .
we compared the isco ( innermost stable circular orbit ) , mbco ( marginally bound circular orbit ) and cpo ( circular photon orbit ) of the said space - time with graphically in comparison with mass - less cases .
moreover , we compute the radius of isco , mbco and cpo for _ extreme _ tn black hole ( bh ) .
interestingly , we show that these _ three radii _ coincides with the killing horizon i.e. the null geodesic generators of the horizon . finally in appendix - a , we compute the center - of - mass ( cm ) energy for tn bh and mass - less tn bh .
we show that in both cases the cm energy is finite . for _ extreme _ nut bh
, we have found that the diverging nature of cm energy . _ first _ , we have observed that a non - asymptotic flat , spherically symmetric and stationary extreme bh is showing such feature . |
a tv broadcast speech database named kalaka-2 @xcite is employed for analysing language - dependent speech .
it was originally designed for language recognition evaluation purposes and consists of wide - band tv broadcast speech recordings ( roughly 4 hours per language ) featuring 6 different languages : basque , catalan , galician , spanish , portuguese and english .
tv broadcast shows were recorded and sampled using @xmath7 bytes at @xmath8 samples / second rate , taking care of including as much diversity as possible regarding speakers and speech modalities .
it includes both planned and spontaneous speech throughout diverse environment conditions , such as studio or outside journalist reports but excluding telephonic channel .
therefore audio excerpts may contain voices from several speakers but only a single language .
for illustrative purposes in figure [ presentacion ] we depict a sample speech waveform amplitude @xmath9 and its squared , semi - definite positive instantaneous energy @xmath2 , respectively . without loss of generality , dropping the irrelevant constants , @xmath10 has units of energy per time .
then , a threshold @xmath3 is defined as the instantaneous energy level for which a fixed percentage of data is larger than the threshold . for instance , @xmath11 is the threshold for which @xmath12 of the data fall under this energy level ( this allows to compare data across different empirical signals @xcite ) .
@xmath3 not only works as a threshold of ` zero energy ' that filters out background ( environmental ) noise , but help us to unambiguously distinguish a speech event , defined as a sequence of _ consecutive _ measurements @xmath4 , from a silence event , whose duration is @xmath1 ( see figure [ presentacion ] ) .
accordingly , speech can now be seen as a dynamical process of energy releases or ` speech earthquakes ' separated by silence events , with in principle different statistical properties given different thresholds @xmath3 . in what follows we address all these properties . + * energy release : a gutenberg - richter - like scaling law in speech * + the energy of a speech event is computed from the integration of the instantaneous energy over the duration of that event @xmath13 where @xmath14 is the inverse of the sampling frequency ( and therefore @xmath5 has arbitrary units of energy ) . in order to get rid of environmental noise , we set a fixed threshold @xmath15 and for each language , we compute its histogram @xmath16 . in figure [ energy ]
we draw , in log - log scales , this histogram for all languages considered ( note that a logarithmic binning was used to smooth out the data ) .
we find out a robust power - law scaling @xmath17 over five decades saturated by the standard finite - size cutoff , where the fitted exponents are all consistent with a language - independent universal behaviour : @xmath18 for spanish , @xmath19 for basque , @xmath20 for portuguese , @xmath21 for galician , @xmath22 for catalan , and @xmath23 for english , all having a correlation coefficient @xmath24 ( for completeness , in the inset of figure [ energy ] we also depict the binned histogram of instantaneous energy @xmath25 for all languages ) . as long as the magnitude in seismicity @xcite is related to the logarithm of the energy release , under this definition @xmath16 can be identified as a new gutenberg - richter - like law in speech .
this may be seen as related to other scaling laws in cognitive sciences @xcite , although at this stage it is still unclear what particular contributions come from both the mechanical ( vocal folds and resonating cavity ) and the cognitive systems .
+ released in the statistics of all languages considered in this study , after a logarithmic binning .
the figure shows a power law @xmath17 , that holds for five decades , truncated for large ( rare ) events by a finite - size cut - off .
( inset panel ) log - log plot of the associated instantaneous energy histogram @xmath25 of each language , which we include for completeness .
we note that this non scaling shape of instantaneous energy is similar to the one found for precipitation rates @xcite.,scaledwidth=40.0% ] * scaling and universality of waiting time distributions . *
+ in a second part , we study on the temporal orchestration of fluctuations , that is , the arrangement of silences or speech interevents of duration @xmath1
. we will pay a special attention to the intraphoneme range ( timescales @xmath26 s @xcite ) , where we assume no cognitive effects are present , in order to focus on the physiological aspects of speech . at this point
we introduce a renormalisation group ( rg ) transformation to explore the origin of temporal correlations .
this technique originates in the statistical physics community and has been previously used in the context of earthquakes @xcite and tropical - cyclone statistics @xcite .
the first part of the transformation consists of a decimation : we raise the threshold @xmath3 .
this in general leads to different interevent distributions @xmath27 .
the second part of the transformation consists in a scale transformation in time , such that renormalized systems become comparable : @xmath28 , @xmath29 , where @xmath30 is the mean interevent time of the system for a particular @xmath3 .
invariant distributions under this rg transformation collapse into a threshold - independent universal curve : an adimensional waiting time distribution .
while the complete fixed point structure of this rg is not well understood yet , recent advances @xcite rigorously found that stable ( attractive ) fixed points include the exponential distribution and a somewhat exotic double power - law distribution , which are attractors for both memoryless and short - range correlated stochastic point processes under the rg flow .
invariant distributions other than the previous fixed points are likely to be unstable solutions of the rg flow , therefore encompassing criticality in the underlying dynamics @xcite .
+ in the left panel of figure [ interevent ] we plot in log - log , for different thresholds , the interevent histogram @xmath27 associated to the english language . in the intraphoneme range , interevents are power - law distributed in every case @xmath31 . in the right panel of the same figure we plot the rescaled histograms @xmath32 ( @xmath33 for every language , yielding a total of 18 curves ) , collapsing under a single curve .
note that the collapse is quite good for those timescales that belong to the intraphoneme range , where only physiological mechanisms are in place , and such collapse is lost for larger timescales .
this suggests that for every language , the statistics are invariant under this rg transformation , and the pattern is robust across languages .
a more careful statistical analysis is required here as the range of the power law , restricted to the intraphoneme regime , is smaller . accordingly , exponent @xmath34 is estimated now following clauset et al.s method @xcite which employs maximum likelihood estimation ( mle ) of a power law model , where goodness - of - fit test and confidence interval are based on kolmogorov - smirnov ( ks ) tests ( comparing the actual distribution with 100 synthetic power law models whose exponent is the one found in the mle to obtain p - values ) .
this method yields a fairly universal exponent @xmath35 within the @xmath36 confidence interval @xmath37 $ ] , ks p - value of @xmath38 and the statistical support for the power - law hypothesis given by a p - value of @xmath39 .
it is well known that the dynamics of speech generation are complex , nonlinear , and certainly poorly explained by the benchmark source - filter theory . the fact that a gutenberg - richter law for the energy release probability distribution during speech emerges opens the possibility of understanding speech production in terms of crackling noise , a highly nonlinear behaviour that was first described in condensed matter physics @xcite and latter found in a variety of natural hazards systems including earthquakes @xcite , rain @xcite or solar flares @xcite to cite some @xcite .
the underlying theory might then describe energy releases as the resonating response function of the system under airflow perturbation ( the so called susceptibility in the statistical physics jargon ) , and the fact that it is a power law distributed quantity is the first evidence of criticality in these systems .
+ the fact that the waiting time distributions are different from the well known stable exponential laws while being invariant under the rg transformation confirms that the system is indeed operating close to a critical point .
note that the currently established low dimensional chaotic hypothesis suffers from being a process with short - range correlations and therefore chaotic speech interevents should typically renormalise into the exponential law .
the most plausible conclusion of this work is that the physiological process of speech production evidences long - range correlations and criticality .
this argument can be put in the same grounds as equilibrium critical phenomena , where only long - range correlations allow the system to escape from the basin of attraction of the trivial rg ( high or low temperature ) fixed points .
as long as the critical solution is an _ unstable _ fixed point of the rg flow , the fact that the underlying dynamics seems to be poised near this point is somewhat remarkable .
if we assume that the dynamics of speech generation at the level of the vocal folds are influenced by the properties of the glottal airflow , then our results would suggest that a similar critical behaviour might take place in other physiological processes involving such airflow , something that has been found empirically in the mechanism of lung inflation @xcite .
+ note also that whereas a simple stochastic processes such as a brownian motion can not explain these results , under the more general paradigm of fractional brownian motion with first return distribution @xmath40 , our findings would be consistent with so called @xmath41 noise ( with hurst exponent @xmath42 ) , on agreement with previous evidence @xcite , this latter being a trait of long range correlations .
this gives further credit to our results , as @xmath41 noise is usually found along with criticality @xcite .
+ although we acknowledge that weakly chaotic systems such as intermittent ones can operate close to criticality , our results suggest that the paradigm of self - organised criticality ( soc ) @xcite - out of equilibrium dissipative systems that self - organise towards a critical state- may be a more adequate modelling scenario than standard low dimensional chaos . if this was to be the case , threshold dynamics @xcite would appear as the essential ingredient that encompasses the nonlinear properties in speech .
these new approaches could further contribute to the development of both ( i ) microscopic ( generative ) models of speech production , and ( ii ) alternative methods of speech synthesis that would profit from the self - similar properties of speech at the intraphoneme range to make refined speech interpolation without needs to incorporate in the synthesis model pieces of real speech .
+ furthermore , the combined fact that ( i ) the results are robust for different human languages and that ( ii ) the timescales involved in this analysis require that the process is purely physiological , lead us to conclude that this mechanism is a universal trait of human beings .
it is an open question if the onset of soc in this system is the result of an evolutionary process @xcite , where human speech waveforms would have evolved to be independent of speaker and receiver distance and perception thresholds . in this context
, further work should be done to investigate whether if similar patterns originate in the communication of other species , and up to which extent other variables , such as ageing , may play a role . +
* acknowledgments . *
the authors would like to thank luis javier rodrguez - fuentes and mikel peagarikano for recording and hand - labeling the speech corpus .
j. neubauer , p. mergell , u. eysholdt , ulrich and h. herzel , spatio - temporal analysis of irregular vocal fold oscillations : biphonation due to desynchronization of spatial modes _ j. acoust .
* 110 * , 3179 - 3192 ( 2001 ) . luis javier rodrguez - fuentes , mikel peagarikano , amparo varona and mireia dez and germn bordel , kalaka-2 : a tv broadcast speech database for the recognition of iberian languages in clean and noisy environments , _ lrec _ , 99105 , ( 2012 ) . | speech is a distinctive complex feature of human capabilities . in order to understand the physics underlying speech production , in this work we empirically analyse the statistics of large human speech datasets ranging several languages .
we first show that during speech the energy is unevenly released and power - law distributed , reporting a universal robust gutenberg - richter - like law in speech .
we further show that such earthquakes in speech show temporal correlations , as the interevent statistics are again power - law distributed .
since this feature takes place in the intraphoneme range , we conjecture that the responsible for this complex phenomenon is not cognitive , but it resides on the physiological speech production mechanism .
moreover , we show that these waiting time distributions are scale invariant under a renormalisation group transformation , suggesting that the process of speech generation is indeed operating close to a critical point .
these results are put in contrast with current paradigms in speech processing , which point towards low dimensional deterministic chaos as the origin of nonlinear traits in speech fluctuations .
as these latter fluctuations are indeed the aspects that humanize synthetic speech , these findings may have an impact in future speech synthesis technologies .
results are robust and independent of the communication language or the number of speakers , pointing towards an universal pattern and yet another hint of complexity in human speech . the description , understanding and modelling of speech is an interdisciplinary topic of current interest for physics @xcite , social and cognitive sciences @xcite , data mining as well as engineering @xcite .
classical speech synthesis technologies and algorithms were firstly based on linear stochastic models and linear prediction @xcite , and their underlying theory , so called source - filter theory of speech production , was initially relying on several key assumptions including uncoupled vocal tract and speech source , laminar airflow propagating linearly , periodic fold vibration , and homogeneous tract conditions @xcite . despite the successes of this benchmark theory , the synthetic speech generated by linear models fails to be ` natural ' . as a matter of fact , current speech synthesizers usually require to incorporate pieces of real speech , e.g. concatenation of smaller speech units ( phoneme and diphone - based synthesis @xcite ) to improve their synthetic output .
+ with the popularisation of nonlinear dynamics , fractals , and chaos theory , the modelling paradigm slightly shifted and nonlinear speech processing emerged @xcite . accordingly , a number of authors pointed towards low dimensional , chaotic phenomena as the underlying mechanism governing the fluctuations in speech .
this paradigm shift has fostered a number of inspiring theoretical vocal - fold and tract models @xcite of increasing complexity , displaying a range of nonlinear phenomena such as bifurcations and chaos @xcite or irregular oscillations @xcite to cite some .
glottal airflow induced by the fluid - tissue interaction has also been modelled using navier - stokes equations @xcite .
nevertheless , the empirical justification for low dimensional chaos is , up to a certain extent , based on preliminary evidence @xcite , and this modelling approach only theoretically justified through an analogy between turbulent states , chaos , and fractals @xcite . as a matter of fact , it has been recently acknowledged in the nonlinear dynamics community that a word of caution should be taken when empirically analysing the low dimensional nature of short , noisy and nonstationary empirical signals , as not only the computation of dynamical invariants such as fractal dimensions or lyapunov exponents is difficult in those cases @xcite , but furthermore correlated stochastic noise can be misleadingly described as to have a low dimensional attractor @xcite .
it is therefore reasonably unclear what modelling paradigm should we follow to pinpoint the nonlinear nature of the fine grained details of speech .
+ in this work we make no a priori assumptions about the underlying adequate dynamical model , and we follow a data driven approach .
we thoroughly analyse the statistics of speech waveforms in extensive real datasets @xcite that extend all the way into the intraphoneme range ( @xmath0 s ) @xcite , enabling us to dissect the purely physiological aspects that are playing a role in the production of speech , from other aspects such as cognitive effects @xcite .
our main results are the following : ( i ) energy releases in speech are power law distributed with a language - independent universal exponent , and accordingly a gutenberg - richter - like law @xcite is proposed within speech .
( ii ) in the intraphoneme range ( @xmath0 s ) , the interevent times ( silences of duration @xmath1 ) between energy releases are also power - law distributed , suggesting long - range correlations in the time fluctuations of the amplitude signal .
( iii ) furthermore , these distributions are invariant under a time renormalisation group ( rg ) transformation , @xcite ) . on the basis of these results ,
we should conclude that the physiological mechanism of speech production self - organises close to a critical point . + of a single - speaker recording ( spanish language ) .
( bottom panel ) instantaneous energy per unit time @xmath2 in an excerpt from the top panel .
the energy threshold @xmath3 , defined as the instantaneous energy level for which a fixed percentage of the entire data remains above that level , helps us to unambiguously distinguish , for a given threshold , speech events ( events for which @xmath4 ) from silence or speech interevents of duration @xmath1 .
the energy released @xmath5 in a speech event is computed from the integration of the instantaneous energy over the duration of that event @xmath6 ( dark area in the figure denotes the energy released in a given speech event).,scaledwidth=45.0% ] |
an electron traversing a path enclosing magnetic flux acquires an additional phase by the magnetic vector potential .
this additional phase causes the aharonov - bohm ( ab ) oscillation of the electronic magnetoconductance@xcite with the period of @xmath3 .
the ab oscillation in solid state devices was first observed by webb _ et al._@xcite and timp _ et al._@xcite in metal and semiconductor ab rings , respectively .
especially , ab rings fabricated on the two - dimensional electron gas ( 2deg ) layer formed in a semiconductor heterojunction structure like gaas / algaas have been studied intensively , because the electron phase coherence in this system extends much longer than the size of an ab ring .
for instance , a structure of a quantum dot embedded in one arm of an ab ring was employed recently for accurate measurements of the phase change of traversing electrons through a quantum dot .
_ et al.__@xcite first demonstrated that the electron transport through a quantum dot embedded in a closed - loop ab interferometer was phase coherent .
however , the phase rigidity , imposed by the onsager relation in the two - terminal geometry of an ab ring , hampered measurements of the genuine phase variation through the quantum dot.@xcite later on , schuster _ _ et al.__@xcite measured the phase evolution via a resonant level of a quantum dot embedded in an open - loop ab interferometer , which allowed multi - terminal conductance measurements , while breaking the time - reversal symmetry of the system . many other phase measurements through various quantum structures
were followed by using the open - loop ab interferometer.@xcite although open - loop ab interferometers are more useful for measuring the phase of the embedded quantum structures , closed - loop ab interferometers are still in often use because of a simpler theoretical interpretation of the results .
detailed studies on closed - loop interferometers , however , often revealed beats in its ab magnetoconductance oscillation .
@xcite beats were first explained in terms of mixing of different multiple transverse sub - band modes arising from a finite width of a ring.@xcite the transverse sub - band modes , with the conductance of @xmath4 per mode , are the quantum states defined by the transverse potential in the interferometer , which can be tuned by the voltages applied to the gates , _
i.e. _ , quantum point contacts .
this model , however , failed to explain the appearance of beats in the conductance with a single transverse sub - band ( stsb ) mode , which was observed by the same group.@xcite later , tan _ et al._@xcite pointed out that the clockwise and the counterclockwise moving electron states in an stsb mode may generate different ab - oscillation frequencies , producing the beats reported in refs .
8 and 10 . in the meantime
, similar beating effect was observed in the ab interferometry using two - dimensional electron gas ( 2deg ) [ two - dimensional hole gas ( 2dhg ) ] systems based on strong spin - orbit interaction ( soi ) materials such as inas and ingaas [ gaas ] heterojunctions.@xcite it was interpreted as the evidence for the revelation of the berry phase @xcite caused by the strong soi . in this case , the fourier spectrum of the beat pattern showed multiple splitting of the @xmath3 peak .
et al._@xcite and meijer _
et al._@xcite showed clear splitting of the @xmath3 peak into multiple peaks in the ensemble - averaged fourier spectrum over several measurements for a fixed and varied gate voltage(s ) , respectively .
they attributed the splitting to the berry - phase @xcite pickup by the traversing electrons due to the soi .
et al._@xcite and yang _
et al._@xcite also observed the multiple splitting of a single fourier - spectrum peak and the evolution of the spectrum for varying magnetic fields .
based on the numerical simulation , the authors also attributed them to the result of the berry - phase pickup .
all these works , however , did not consider the effect of possible mixing of multiple conducting channels . in this study , we investigated the behavior of the magnetoconductance from a closed - loop ab interferometer fabricated on a gaas / al@xmath0ga@xmath1as heterostructure 2deg , with a very weak soi .
the magnetoconductance always revealed clear beats of ab oscillations .
the corresponding fourier spectrum exhibited multiple peaks closely spaced around the field value where the @xmath3 peak was expected .
the patterns were very similar to those observed previously from the ab rings fabricated on strong soi materials . to interpret our observation of the beats in our system with a very weak soi , we simulated two - dimensional ( 2d ) ab ring using the method adopted by tan _
_ et al.__@xcite in the simulation , even in a fixed transverse sub - band mode , additional eigen - modes existed to contribute to the electron transport as a function of the fermi level and a magnetic field .
for the remainder of this paper these modes will be referred as longitudinal modes .
these modes were found to evolve in different ways depending on the geometry and thus induce beats of ab oscillations in a 2d interferometer .
we were able to explain all our results in terms of the formation of the 2d multiple longitudinal modes in an stsb without resorting to the strong - soi effect .
in fact , measurements of the weak localization ( wl ) and the shubnikov - de haas ( sdh ) oscillation in our system revealed that the soi was almost negligible . in our observation
the @xmath2 periodicity was noticeable only around the nodes of the beats .
in addition , the nodes of the beats formed a parabolic distribution for varied voltages applied to one of the loop - forming side gates , which was not observed in the previous studies.@xcite these two general features of our data were in accordance with the prediction of the 2d multiple - longitudinal - modes effect in an stsb .
figure 1 shows a gate - defined closed - loop ab interferometer , fabricated on a wafer of gaas / al@xmath0ga@xmath1as heterostructure 2deg .
the electron density of the 2deg was about @xmath5 @xmath6 with the mobilities higher than @xmath7 @xmath8 v@xmath9s@xmath9 at 4.2 k. the 2deg layer resided 70 nm below the surface of the wafer .
the interferometer was defined by an island gate , four side gates , and two quantum - dot gates as shown in the figure .
the shape of this structure is similar to the one reported previously.@xcite the electron path inside the interferometer was defined by tuning negative voltages to these gates . unlike a closed - loop interferometer defined by the mesa formation ,
@xcite the number of transverse conducting channels ( sub - bands ) in each ( left or right ) arm could be fine - tuned by the six side gates .
this was a big advantage over mesa - defined ab rings , where the existence of multiple transverse sub - bands was inevitable in practice . in our study , the number of transverse sub - bands for each path was kept close to or below one to avoid any possible interference among electrons from different transverse sub - bands .
the quantum dot embedded in the interferometer in fig .
1 was not formed during measurements of beats , despite the physical presence of the gates to form it in the sample .
all the measurements were made at the base temperature of 10 mk in a dilution fridge . the corresponding electron temperature estimated from measurements of the coulomb blockade in the quantum dot @xcite was around 140 mk .
the standard phase - sensitive lock - in technique was employed to improve the signal to noise ratio .
the ab magnetoconductance oscillations were taken by observing the current between the two ends ( the source and the drain ) of the ab ring in an excitation voltage of 10 @xmath10v rms . for measurements of the wl and the sdh oscillation ,
a four - probe configuration was employed , where a 15-na rms current was applied between the source and the drain while monitoring the corresponding voltage difference .
the magnetoconductance of the interferometer at various voltages of side - gate 1 ( @xmath11 ) was measured by applying a 10-@xmath10v - rms excitation voltage to the interferometer , while monitoring the source - drain current .
the magnetic field was swept in the range between -1500 g and 1500 g , while the @xmath11 was varied from -271.35 mv to -291.15 mv at intervals of 1.65 mv . since the conductance for any @xmath11 was lower than
@xmath4 it was assumed that less than an stsb @xcite was open for each path of the interferometer .
since the visibility of the ab oscillation was about 2% the background current level was subtracted for clarity . as shown in fig .
2(a ) the field dependence of the conductance exhibits clear beats of ab oscillations for any values of @xmath11 .
in addition , the phase evolved discontinuously in the variation range of @xmath11 .
this phenomenon of the rigid phase has been observed by others@xcite and interpreted by the onsager relation in a two - terminal closed - loop interferometer.@xcite the relation stipulates that the electronic conductance through a closed - loop interferometer should be symmetric with respect to the @xmath12=0 axis , allowing only 0 or @xmath13 phase . as a result , the phase change occurs only by @xmath13 via a double - frequency ( @xmath2 period ) ab oscillation regime.@xcite the expected phase change in a one - dimensional ( 1d ) closed - loop interferometer , with an stsb per path without considering finite width of the electron path , is shown in fig .
2(b ) . as in the figure
, the phase stays unaltered for the variation of the fermi energy ( @xmath14 ) in the leads until it suddenly changes by @xmath13 .
the simulation also suggests that the frequency of the ab oscillation doubles with @xmath2 periodicity between the transition of the two phase states . for a 1d closed - loop interferometer , with an stsb in each arm
, this transition should occur at the same @xmath14 in all magnetic fields .
in other words , as in fig .
2(b ) , the transition region should be in parallel with the @xmath12-field axis . in our measurements
, however , the transition region distributes rather parabolically in the @xmath11-vs-@xmath12-field plot [ fig .
these features were reproducible even after thermal recycling .
10 and 14 report the existence of @xmath2 period in the beats of ab oscillations , but without this parabolic distribution .
similar behavior was observed with the variation of the source - drain bias voltage .
a parabolic distribution of the transition regions was also observed when the voltage of the island gate was varied , while fixing the voltages for all the other gates used to form the interferometer .
such a beating effect without the parabolic distribution of the @xmath2 transition region mentioned above has been observed in various systems like metallic ab rings,@xcite gaas / algaas 2deg ab rings,@xcite gaas / algaas 2dhg rings,@xcite inas - based@xcite and ingaas - based@xcite 2deg rings .
although the beating features are similar to each other , interpretations of the origin of the beats are varied .
two main models suggested to explain the beats are based on picking up the berry phase induced by the soi and mixing of multiple conducting channels caused by the finite width of an ab ring involved .
the berry - phase - related interpretation of the beats is favored for the ab rings formed on the relatively strong soi systems like gaas / algaas - based 2dhg , and inas - based and ingaas - based 2deg s .
in contrast , the model of mixed multiple conducting channels is favored for ab rings in metallic films and gaas / algaas - based 2deg s with relatively small soi.@xcite to estimate the strength of the soi in our system , both the wl and the sdh oscillation were measured on an identical sample that was used to measure the beats .
fitting the anti - localization dip in the magnetoresistance for strong soi materials leads to resolving the soi strength.@xcite by contrast , the soi gives two different frequencies for the sdh oscillation .
it results in beats in the sdh oscillation , allowing one to estimate the soi strength .
@xcite figs .
3(a ) and 3(b ) illustrate the magnetoresistances of the wl and the sdh oscillation , respectively , measured in a hall - bar geometry formed on our sample .
all the gates in the interferometer were grounded during the measurements .
3(a ) shows the magnetoresistance averaged over 37 measurements with the field resolution of 0.2 g , which exhibits no discernible anti - localization dips .
3(b ) shows the measured sdh oscillation , without any beats , either .
these results strongly imply that the soi was very weak or almost negligible in the gaas / al@xmath0ga@xmath1as heterostructure 2deg of our interferometer .
although it is known that metallic gates may enhance the strength of the soi , the enhancement is usually less than 30@xmath15 of the original value.@xcite we thus expect that the soi caused little effect on our magnetoconductance measurements . in our measurements , only an stsb was set to open in each arm .
the beats of ab oscillation for the stsb conduction were experimentally observed previously,@xcite where the beats were theoretically interpreted in terms of mixing of clockwise and counterclockwise moving states of electrons with different frequencies @xcite in an stsb mode .
we employed the model as used in ref . 9 to simulate the beats and the evolution of their transition region as a function of @xmath11 .
the result indicates that the multiple longitudinal modes existing in an stsb mode generate the ab oscillations .
however , the multiple longitudinal modes give more than two oscillation frequencies , with the number of the oscillation frequencies varying with the magnetic field , due to the finite width of the electron path inside the interferometer .
the magnetic field dependence will be discussed in the next section in more detail .
we adopted the same form of shrdinger equation as used in ref .
9 , but with varied parameter values to fit our experimental conditions .
the eigenvalues are @xmath16 where @xmath17 , @xmath18 , @xmath19 , @xmath10 is the effective mass of the electron , @xmath20 is the transverse confinement strength of the gates defining the electron path , and @xmath21 is the average radius of the interferometer . here ,
the quantum number @xmath22 is the transverse sub - band index and @xmath23 is related with the longitudinal motion.@xcite in our measurements , we tuned the interferometer to have an stsb , corresponding to @xmath22=0 . in fig .
4(a ) , the energy band @xmath24 is shown for an stsb mode with @xmath25 and @xmath2630 . as shown in fig .
4(a ) , several conducting modes exist even for an stsb mode because of the existence of multiple longitudinal modes .
the corresponding energy eigenvalues depend on @xmath23 as well as @xmath22 . for this dependence ,
the dispersion of the bands is asymmetric about the band minima .
the minima shift to higher energies for nonzero magnetic fields,@xcite which result in breaking of the discrete symmetry under the translation along @xmath12-axis .
as indicated by filled ( open ) circles in the figure , this makes band crossing ( splitting ) points respond nonlinearly to the applied magnetic field .
4(b ) is an expanded band diagram of the region in the dashed box in fig .
the band - splitting points show a clear downturn curvature as represented by open circles , the shape of which is very similar to that of the conductance with the @xmath2 period in fig .
the conductance for a given fermi energy @xmath14 in the dashed - box region in fig .
4(a ) was calculated using the landauer - bttiker formula@xcite @xmath27 ^ 2 + { \gamma}^2},\end{aligned}\ ] ] where only @xmath25 is taken into account and the transmission probability via an eigen mode is given by the breit - wigner formula at zero temperature . here , @xmath28 is the coupling constant between the emitter ( or the collector ) and the ab ring .
the @xmath28 was set at 3% of @xmath29 , which corresponds to a weak coupling regime of the system .
the calculated results are shown in fig .
the conductance shows a clear beat pattern with a variation of the envelope as a function of @xmath14 .
5(b ) shows similar behavior observed experimentally for varied @xmath11 .
the figures indicate that the @xmath2 ( @xmath3 ) oscillation predominates in the nodes ( antinodes ) of the beats .
the region of dominant @xmath2 oscillation appears in different magnetic fields as @xmath14 is varied .
a grey - scale plot of the calculated conductance is shown in fig 6 . the @xmath14 was set to the range in the dashed square box in fig .
the band shown in fig .
4(b ) is overlapped on the grey - scale plot by white solid lines .
contrary to the results calculated for an stsb in a 1d interferometer as in fig .
2(b ) , the region where the nodes of the beats appear shows an @xmath2 oscillation and distributes parabolically in an @xmath14-vs-@xmath12-field plot .
this result is in qualitative agreement with our observation shown in fig .
clearer revelation of the double frequency region in the simulation may be from smaller @xmath28 value and higher visibility of ab oscillations than those in the measurement .
the parabolic evolution of the node region can be naively understood with the band diagram shown in fig .
4 . according to eq .
( 2 ) the overall conductance through an ab ring is determined mainly by the bands located close to @xmath14 .
hence , if @xmath14 is at a crossing point of two bands ( marked by a filled circle in the figure ) the @xmath3 oscillation becomes predominant .
by contrast , the @xmath2 oscillation predominates if @xmath14 is at a band splitting point marked by an open circle in the figure .
the conductance itself is bigger for @xmath3 oscillation , since two longitudinal modes simultaneously contribute to the conductance , while a single longitudinal mode contributes to the conductance for @xmath2 oscillation .
this is the reason why @xmath3 ( @xmath2 ) oscillation is dominant in the antinodes ( nodes ) . since these band crossing ( splitting )
points evolve parabolically as a function of @xmath14 and @xmath12 , the antinodes ( nodes ) also evolve in the same fashion .
it should be emphasized that the finite width of the electron path in an ab interferometer is responsible for the parabolic evolution of the band crossing points . according to our simulation , however , a parabolic evolution with an upturn curvature is also possible in different regions of @xmath14 ( not shown ) .
similar experimental result was also observed in a different region of the gate voltages . the beat feature similar to fig .
5 was reported in mesa - defined closed - loop ab interferometers,@xcite with the results interpreted in terms of 2d multiple - longitudinal - modes effect@xcite in an stsb . in these works , however , no parabolic evolution of the beat pattern was reported .
the parabolic evolution gives an additional confirmation of the validity of the model and better understanding of the 2d multiple - longitudinal - modes effect in an stsb .
observing the smooth evolution of the node region was possible in our study , because the number of transverse sub - bands was fine - controlled by the six individual gates , rather than by a single gate covering the whole mesa - defined ab ring as used in the other measurements .
we tuned all the gates individually to leave an stsb open .
only one gate was then slightly modified to observe such an evolution .
the fourier transform analysis is a very effective way to analyze the beating of ab oscillation induced both by the multiple - conducting - channel effect@xcite and the soi effect.@xcite the fourier spectrum gives the information on different conducting channels ( or modes ) inside an interferometer with very small frequency differences . in this section
, we present the fourier spectrum of two representative beats , which show the antinodes and the nodes at @xmath12 = 0 .
the zero - padding technique@xcite , that was used to analyze the beats induced by the soi effect,@xcite was adopted for our fourier transform analysis .
the magnetoconductance data taken in the range between -1500 g and 1500 g were used for the analysis , because sdh oscillation started to appear at around @xmath30 2000 g due to the high mobility of electrons in the sample .
within this range , four different widths of the hamming window@xcite ( @xmath30500 g , @xmath30900 g , @xmath301200 g and @xmath301500 g ) were used .
this allowed a close examination of any field - dependent effect .
the left ( right ) panel of fig .
7(a ) is the fourier spectrum for the antinode ( node ) at @xmath12=0 .
these two cases show different features , but their magnetic field dependencies are similar to each other . when the antinode is at @xmath12=0 , the fourier spectrum shows a main center peak and a trace of small side peaks in a low magnetic field range . with increasing magnetic field ,
the main center peak starts to split , while the side peaks become clearer .
when the node is at @xmath12=0 , however , the amplitudes of the main center peak and symmetric side peaks are comparable even in a low - magnetic - field range . with increasing magnetic field in this case , the center peak starts to split , while the amplitudes of the side peaks increase .
7(b ) is the fourier spectrum of the calculated magnetoconductance when an antinode and a node are at @xmath12=0 .
the features are similar to those of the experimental fourier spectrum .
the evolution of the fourier spectrum as a function of the magnetic field can be naively explained as follows .
first , the band crossing points marked by filled circles in the figure 4(a ) give a fourier - transform peak ( the center peak ) .
second , open circles marked in the figure give two fourier - transform peaks ( the side peaks ) , where one is slightly lower and another is slightly higher than the peak given by the band crossing points .
these are major three peaks shown in the spectrum .
the frequency difference and the field dependence of the peaks are caused by the breaking of the discrete symmetry under the translation along the @xmath12 axis .
even with many energy bands around @xmath14 , the conductance is mainly determined by the bands close to @xmath14 .
thus , if the data showing an antinode at @xmath31 are taken for the fourier spectrum analysis with a very narrow magnetic - field window only a center peak appears in the spectrum . as the width of the magnetic field window increases ,
the nodal part is included in the analysis to produce the side peaks in the spectrum . by the same analogy ,
if the data showing node at @xmath31 are taken for the analysis with a narrow magnetic field window only the side peaks appear in the spectrum . as the antinodal part
is included in the analysis by enlarging the magnetic field window the center peak appears in the spectrum .
these features are illustrated in fig .
however , the experimental data given in the right panel of fig .
7(a ) show three peaks ( a center peak and two side peaks ) even in a low - magnetic - field range .
that is because the magnetic field window is wide enough to contain the antinodes as well as the node as shown in fig .
. similar evolution of fourier - spectrum peaks as a function of magnetic field was also reported in refs .
13 and 14 . those were interpreted as evidences for the berry phase pickup by the soi , although the interpretation in ref .
13 has been controversial.@xcite we believe that the evolution of peaks observed in the fourier spectrum as a function of magnetic fields are not a conclusive evidence of the spin - orbit coupling,@xcite since the same effect is also observable in systems with negligible spin - orbit coupling .
the beats in this case can be explained by the 2d effects inside the interferometer as clearly demonstrated in our study , both in experiment and in simulation .
meanwhile , the center - peak splitting and the side peaks were also observed in other magnetoconductance measurements using ab rings fabricated on strong soi materials.@xcite in these works , the splitting of the center peak was focused and regarded as an evidence for the berry - phase pickup by the soi .
it was pointed out , however , that the splitting was too large to be explained by the berry phase pickup only . in the works ,
the existence of the side peaks was not clearly explained , either .
as shown in our data , the splitting of the center peaks in fig .
7(a ) is around 0.001 g@xmath9 , which is comparable to those in refs .
11 and 12 .
even with almost negligible soi in our system , the features of the fourier spectrum of the measured magnetoconductance show a similarity to those found in strong - soi systems.@xcite our study demonstrates that the features in the fourier spectrum can be generated irrespective of the berry - phase pickup .
the beats of ab oscillations in a closed - loop interferometer fabricated on gaas / al@xmath0ga@xmath1as 2deg were observed . unlike mesa - defined interferometers , in our measurements ,
the electron path in each arm of the interferometer was kept less than a single transverse sub - band ( stsb ) by the fine control of six individual path - forming side gates .
this enabled us to study the effect of the two - dimensional longitudinal modes developing in an stsb induced by the external magnetic field .
our simulation indicates that the experimentally observed beats of ab oscillations resulted from the multiple oscillation frequencies induced by finite width of the interferometer even with an stsb mode .
the @xmath2 ab oscillations , imposed by the onsager relation in our closed - loop two terminal geometry , appeared in the nodes of beats , while usual @xmath3 ab oscillations appeared in the antinodes , which was in agreement with our simulation results .
it was also found that nodes of the beats distributed parabolically with varying the voltages of the side gates .
we confirm that the phenomena can be explained by the 2d multiple - longitudinal - modes effect in the presence of an stsb , even without resorting to the strong spin - orbit interaction .
this work was supported by electron spin science center in pohang university of science and technology ( postech ) and pure basic research grant r01 - 2006 - 000 - 11248 - 0 administered by kosef .
this work was also supported by the korea research foundation grants , krf-2005 - 070-c00055 and krf-2006 - 312-c00543 , and by postech core research program .
y. aharonov and d. bohm , phys . rev . *
115 * , 485 ( 1959 ) .
r. a. webb , s. washburn , c. p. umbach , and r. b. laibowitz , phys .
lett . * 54 * , 2696 ( 1985 ) .
g. timp , a. m. chang , j. e. cunningham , t. y. chang , p. mankiewich , r. behringer , and r. e. howard , phys .
lett . * 58 * , 2814 ( 1987 ) .
a. yacoby , m. heiblum , d. mahalu , and hadas shtrikman , phys .
lett . * 74 * , 4047 ( 1995 ) .
a. yacoby , r. schuster , and m. heiblum , phys
b * 53 * , 9583 ( 1996 ) .
r. schuster , e. buks , m. heiblum , d. mahalu , v. umansky , and h. shtrikman , nature * 385 * , 417 ( 1997 ) .
yang ji , m. heiblum , d. sprinzak , d. mahalu , and hadas shtrikman , science * 290 * , 779 ( 2000 ) ; yang ji , m. heiblum , and hadas shtrikman , phys .
88 * , 076601 ( 2002 ) ; a.w .
holleitner , c.r .
decker , h. qin , k. eberl , and r. h. blick , phys .
lett . * 87 * , 256802 ( 2001 ) .
j. liu , w. x. gao , k. ismail , k. y. lee , j. m. hong , and s. washburn , phys .
b * 48 * , 15148 ( 1993 ) .
tan and j. c. inkson , phys .
b * 53 * , 6947 ( 1996 ) .
a. a. bykov , z. d. kvon , and e. b. olshanetskii , proc .
compound semiconductors , pp .
909 - 914 ( 1995 ) .
a. f. morpurgo , j. p. heida , t. m. klapwijk , b. j. van wees , and g. borghs , phys . rev
. lett . * 80 * , 1050 ( 1998 ) .
f. e. meijer , a. f. morpurgo , t. m. klapwijk , j. nitta , and t. koga , phys .
b * 69 * , 035308 ( 2004 ) .
jeng - bang yau , e. p. de poortere , and m. shayegan , phys .
88 * , 146801 ( 2002 ) .
m. j. yang , c. h. yang , and y. b. lyanda - geller , europhys . lett . *
66 * , 826 ( 2004 ) .
m. v. berry , proc .
london , ser .
a , * 392 * , 45 ( 1984 ) .
u. meirav , m. a. kastner , and s. j. wind , phys .
lett . * 65 * , 771 ( 1990 ) .
g. cernicchiaro , t. martin , k. hasselbach , d. mailly , and a. benoit , phys .
lett . * 79 * , 273 ( 1997 ) ; s. pedersen , a. e. hansen , a. kristensen , c. b. srensen , and p. e. lindelof , phys .
b * 61 * , 5457 ( 2000 ) ; r. hussler , e. scheer , h. b. weber , and h. v. lhneysen , phys . rev .
b * 64 * , 085404 ( 2001 ) . a. d. stone , phys
* 54 * , 2692 ( 1985 ) ; s. washburn and r. a. webb , rep .
. phys . * 55 * , 1311 ( 1992 ) .
p. d. dresselhaus , c. m. a. papavassiliou , r. g. wheeler , and r. n. sacks , phys
lett . * 68 * , 106 ( 1992 ) ; c. kurdak , a. m. chang , a. chin , and t. y. chang , phys . rev .
b * 46 * , 6846 ( 1992 )
. g. l. chen , j. han , t. t. huang , s. datta , and d. b. janes , phys .
b * 47 * , 4084 ( 1993 ) ; s. pedersen , c. b. srensen , a. kristensen , p. e. lindelof , l. e. golub , and n. s. averkiev , phys .
b * 60 * , 4880 ( 1999 ) ; t. koga , j. nitta , t. akazaki , and h. takayanagi , phys .
* 89 * , 046801 ( 2002 ) ; j. b. miller , d. m. zumbhl , c. m. marcus , y. b. lyanda - geller , d. goldhaber - gordon , k. campman , and a. c. gossard , phys .
lett . * 90 * , 076807 ( 2003 ) .
j. luo , h. munekata , f. f. fang , and p. j. stiles , phys .
b * 38 * , 10142 ( 1988 ) ; j. luo , h. munekata , f. f. fang , and p. j. stiles , phys . rev .
b * 41 * , 7685 ( 1990 ) ; p. ramvall , b. kowalski , and p. omling , phys . rev .
b * 55 * , 7160 ( 1997 ) ; w. yang and k. chang , phys .
b * 73 * , 045303 ( 2006 ) .
j. nitta , t. akazaki , h. takayanagi , and t. enoki , phys .
. lett . * 78 * , 1335 ( 1997 ) .
m. bttiker , y. imry , r. landauer , and s. pinhas , phys .
b * 31 * , 6207 ( 1985 ) .
s. k. mitra , _ digital signal processing : a computer - based approach _ ( mcgraw - hill / irwin , boston , 2001 ) .
a. v. oppenheim , r. w. schafer , and j. r. buck , _ discrete - time signal processing _
( prentice hall , upper saddle river , n.j . , 1999 ) .
apoorva g. wagh and veer chand rakhecha , phys .
lett . * 90 * , 119703 ( 2003 ) .
a. g. malshukov and k. a. chao , phys .
* 90 * , 179701 ( 2003 ) . in our measurements , we first checked separately that the paths between each of six gates ( side gates 1,2,3,4 , and quantum - dot gates 1,2 in fig . 1 ) and the island gate ( in fig .
1 ) gave a single transverse sub - band mode by tuning each corresponding conductance below @xmath4 .
we then checked the conductance through the interferometer as a whole to be lower than @xmath4 .
this confirmed us that the interferometer kept only a single transverse sub - band mode .
the resulting conductance with the subtraction of the background revealed clear ab oscillations with beats even for a low visibility .
* figure captions * + + figure 1 .
sem image of the gate - defined closed - loop aharonov - bohm interferometer fabricated on a gaas / al@xmath0ga@xmath1as heterostructure two - dimensional electron - gas layer .
+ + + figure 2 .
( color ) ( a ) the magnetoconductances for various voltage to side - gate 1 , @xmath11 , after subtracting the smooth backgrounds from the raw data , as a function of magnetic fields .
the magnetic field was varied from -500 g to 500 g , while @xmath11 from -271.35 mv ( top ) to -291.15 mv ( bottom ) at intervals of 1.65 mv .
( b ) the phase rigidity expected in a one - dimensional closed - loop ab - ring is illustrated .
( c ) the two - dimensional pseudocolor plot of the magnetoconductance shown in ( a ) .
a double frequency regime is guided by the dashed line .
for both pseudocolor plots , the red ( blue ) color represents the high ( low ) magnetoconductance .
+ + + figure 3 .
( a ) normalized magnetoresistance averaged over 37 independent measurements and ( b ) the longitudinal hall resistance of a hall - bar structure , both plotted as a function of magnetic fields .
+ + + figure 4 .
( color online ) ( a ) the calculated energy - band diagram of a two - dimensional ab ring .
the energy eigenvalues for @xmath320 and @xmath3330 modes are plotted as a function of magnetic fields .
multiple bands correspond to different @xmath23 values .
the filled circles denote the band - crossing points , while the open circles denote the middle point between two neighboring band - crossing points .
( b ) the band diagram of the dashed - box region in ( a ) , enlarged for clarity .
+ + + figure 5 .
( a ) the magnetoconductance calculated for @xmath14=0.306 mev and 0.298 mev , where the fermi energies were set to lie in the lowest sub - band to simulate the measurement condition .
( b ) measured magnetoconductance for @xmath11=-274.65 mv and -287.85 mv .
+ + + figure 6 .
( color online ) grey - scale plot of the conductance , calculated for a two - dimensional electron - gas ab - ring , as a function of @xmath14 and magnetic field , where the band shown in fig .
4(b ) is overlapped in white solid lines .
the white dashed line is a guide to denote the region where the @xmath2 oscillation is dominant .
+ + + figure 7 .
( color online ) ( a ) the fourier spectrum of the @xmath3 magnetoconductance measured at different @xmath11 in a few magnetic - field windows .
in the left ( right ) panel , with @xmath11 = -289.50 mv ( -277.95 mv ) , the antinode ( node ) of beats was at @xmath12=0 . in the left ( right ) panel ,
the magnetic field was varied in the ranges of @xmath30500 g ( @xmath30500 g ) , @xmath30900 g ( @xmath301200 g ) , and @xmath301200 g ( @xmath301500 g ) , indicated by the thick solid line , the thin solid line , and the dotted line , respectively .
( b ) the fourier spectrum of @xmath3 magnetoconductance calculated for different values of @xmath14 and a few magnetic - field windows are shown for comparison . in the left ( right )
panel , with @xmath14=0.297 mev ( 0.305 mev ) , the beating features for @xmath12=0 are similar to those of ( a ) .
in the left ( right ) panel , the magnetic field was varied in the ranges of @xmath30800 g ( @xmath30800 g ) , @xmath301200 g ( @xmath301600 g ) , and @xmath301600 g ( @xmath302400 g ) , indicated by the thick solid line , the thin solid line , and the dotted line , respectively . | one of the points at issue with closed - loop - type interferometers is beating in the aharonov - bohm ( ab ) oscillations .
recent observations suggest the possibility that the beating results from the berry - phase pickup by the conducting electrons in materials with the strong spin - orbit interaction ( soi ) . in this study , we also observed beats in the ab oscillations in a gate - defined closed - loop interferometer fabricated on a gaas / al@xmath0ga@xmath1as two - dimensional electron - gas heterostructure .
since this heterostructure has very small soi the picture of the berry - phase pickup is ruled out .
the observation of beats in this study , with the controllability of forming a single transverse sub - band mode in both arms of our gate - defined interferometer , also rules out the often - claimed multiple transverse sub - band effect .
it is observed that nodes of the beats with an @xmath2 period exhibit a parabolic distribution for varying the side gate .
these results are shown to be well interpreted , without resorting to the soi effect , by the existence of two - dimensional multiple longitudinal modes in a single transverse sub - band .
the fourier spectrum of measured conductance , despite showing multiple @xmath3 peaks with the magnetic - field dependence that are very similar to that from strong - soi materials , can also be interpreted as the two - dimensional multiple - longitudinal - modes effect . |
the all - sky survey high resolution air - shower detector ( ashra ) @xcite is an experiment for obtaining fine images of air - showers produced by very high energy ( vhe ) cosmic - ray particles as well as directly observing starlight to monitor optical transients .
ashra has an potential to provide a systematic exploration of extragalactic vhe particle radiators in the universe .
the ashra observational station consists of 12 light collectors covering 77% of the entire sky with a total of 50 mega pixels in cmos sensor arrays .
each light collector has a 42@xmath0 field of view ( fov ) and a spot size resolution of a few arcmin .
the focal surface of the ashra optical system consists of a photoelectric lens image intensifier tube ( pli ) sensitive to uv and visible rays .
the pli reduces an image to the size of a solid state imaging device , such as a cmos sensor , using the electric lens effect .
since this feature drastically reduces the pixel cost compared to experiments using photomultiplier tube ( pmt ) arrays @xcite , it is a key concept in the ability of ashra to perform all - sky surveys with a few arcmin resolution .
details of the design and performance of pli are described in this paper .
the pli is a 1st generation image intensifier tube , sensitive to uv and visible photons , fabricated by toshiba electron tubes & devices co. , ltd ( tetd ) .
[ fig : photo ] shows a photograph of a 20-in .
pli .
the pli features are as follows : 1 . the worlds largest image intensifier , 2 .
large input diameter of 500 mm , 3 .
small output diameter of @xmath125 mm , 4 .
high imaging capability of 2 lp / mm at input window , 5 .
very fast response using a p47 phosphor screen , 6 .
good single photoelectron detection capability , 7 .
photon - to - photon amplification factor of @xmath27 .
the 20-in .
pli is based on the success on the development of a 16-in .
uv - ray image intensifier tube @xcite , which is a modification of a commercial x - ray image intensifier widely used in medical imaging .
image intensifier tubes were already used in our observation using prototype detectors @xcite .
the pli consists of three main components ; a 20-in .
input window , main parts of vacuum tube containing several electrodes , and an output window .
the input window is a uv transparent glass , sensitive to uv and visible photons ( not a thin metal plate , as is used in x - ray image intensifiers ) .
the input window has a spherical shape with the proper curvature radius to work as the focal surface of the ashra light collector .
photoelectrons generated at a photocathode on the inner surface of the input window are drawn toward the output window by an electric field produced by electrodes .
the voltage applied to each electrode was optimized to get a uniform resolution over the whole effective area of pli .
the anode potential of the electric lens is @xmath240 kv .
the input image is reduced by a factor of 20 or more as a result of the electric lens effect .
the concave - shaped output focal surface of the electric lens is formed of a phosphor material screen on a concave - shaped fiber optic plate ( fop ) . as a phosphor material we used y@xmath3sio@xmath4:ce , known as p47 , to take advantage of its very fast 10% decay constant of @xmath2110 ns .
an incident photoelectron is amplified to @xmath290 photons at the phosphor screen .
finally , a planar image is obtained on the output side of the fop window .
ashra-1 collaboration already started the observation using some of the finished 20-in .
plis as reported in refs . @xcite .
qualitatively , similar results to the performance described in this paper were obtained with plis used in the observation . here , we quantitatively evaluated various performance parameters using the latest pli : the quantum efficiency , phosphor gain and imaging capability . in the following
, each item is described in detail .
at first , we evaluated the sensitivity of the pli .
[ fig : qe ] shows the quantum efficiency as a function of photon incident position ( where the error bars represent the quadratic sum of the statistical and systematic errors ) .
the quantum efficiency was calculated as a sensitivity relative to a calibrated pmt .
a uv led with a wavelength peaked at 365 nm was used as a light source .
first , we measured the photocathode current of the calibrated pmt where the voltage of 250 v was applied between photocathode and dynodes .
all the dynodes and the anode were connected and resultant current saturated at much lower voltage .
second , we measured the photocathode current of the pli where all the electrodes and the anode were connected and the voltage of 250 v was applied between photocathode and electrodes .
we confirmed that the measured current was almost saturated at 250 v. since the obtained current ratio corresponds to the ratio of the quantum efficiencies , the quantum efficiency of the pli was calculated for each input position .
the dominant uncertainty in the quantum efficiency was the uncertainty in the measurements of the quantum efficiency for the calibrated pmt . as shown in fig .
[ fig : qe ] , the quantum efficiency was measured to be 7.7% at the center of the input window and was increasing toward the edge . at the edge
, the quantum efficiency was measured to be 10.5% .
the sensitivity was confirmed using pulsed yag laser ( peak wavelength of 355 nm ) ; we obtained the quantum efficiency consistent with the results described above at the center region . in order to evaluate the performance of the phosphor screen
, we investigated the single photoelectron response . to do this
, we used a led illuminated by pulsed input voltage .
the output photons from led was set to 0.1 photoelectrons per pulse on average , for which we could neglect a signal of two photoelectrons or more . as a result
, we obtained a pure one photoelectron signal .
the gain properties of the output screen were measured with a pmt directly coupled to the output fop of the pli .
[ fig : spedst ] shows the photon distribution emitted from the output screen for single photoelectron incidence .
the hatched area corresponds to the pedestal distribution .
we found that about 90 photons were emitted on average .
an acceptable structure was obtained for the single photoelectron distribution . to evaluate photon - to - photon gain of pli , on the other hand
, we used a very stable yap pulser ( yalo@xmath5:ce+241am , peak wavelength of 370 nm , and @xmath6 decay time of 28 ns ) . at first
, photon intensity of the yap pulser was measured by the calibrated pmt .
then , yap pulser was placed at the several positions of the input window of the pli .
the output photons were measured with the same pmt directly coupled to the output window of the pli .
the resultant photon - to - photon gains were obtained as the ratio of the pmt signals after correcting sensitivity difference between yap and p47 spectrum , and are plotted in fig .
[ fig : p2p ] as a function of the photon incident position .
the error bars represent the quadratic sum of the statistical and systematic errors . the most significant contribution to the overall error
was the uncertainty in the p47 spectrum , which might cause the sensitivity difference of the calibrated pmt .
the obtained photon - to - photon gain includes the following effects : 1 .
input glass transmittance , 2 .
quantum efficiency , 3 .
collection efficiency of photoelectrons , 4 .
efficiency of the output phosphor screen , 5 . averaged photon output per single photoelectron incidence . in the previous section ,
the measured quantum efficiency only accounts for the input glass transmittance and the raw quantum efficiency ( points 1 and 2 above ) , which is the same definition as the general use . in photon - to - photon gain measurements ,
however , the efficiency of the photoelectron collection and of the output screen are also included .
combined with the single photoelectron distribution measurement shown in fig .
[ fig : spedst ] , we confirmed that these efficiencies ( points 3 and 4 above ) are reasonably high .
note , however , that there are several deficits in the peripheral region of the output phosphor screen of the pli tested in this paper .
the brightness of the phosphor screen in those deficits was much worse ( 1/2 or less ) than that of normal area .
the radial scan of the photon - to - photon gain was performed along the direction where no such deficits in the output phosphor screen was seen in order to demonstrate the normal performance expected in the pli design .
[ fig : decay ] shows an typical waveform of the yap signal taken through the pli .
it represents the convolution of two decay times of the yap and the p47 phosphor screen . from the fit to the data shown as the long dashed line in fig .
[ fig : decay ] where the yap decay time was fixed to 28 ns ( @xmath6 ) , the 10% decay time of the p47 screen was determined to be 110 ns , which is consistent with the typical catalog value . this very fast decay feature enables us to detect cosmic - ray air - light signals with a high signal - to - noise ratio .
the point spread function ( psf ) was next measured to evaluate the pli resolution quantitatively .
a uv - blue laser was incident on the pli through a 10-@xmath7 m pinhole and a focusing lens , and the output image was taken by an 6-mega - pixel digital camera .
the laser was mounted on a 2-axis positioning stage such that its direction was parallel to the pli axis , where the pli axis is defined as a line between the center of the input window and the center of the output window . by varying
the laser incident position in the plane perpendicular to the pli axis , we measured the pli resolution as a function of photon incident position .
the spot size of the laser was measured before evaluating the pli resolution .
we found that the full width half maximum ( fwhm ) of the laser spot was within 120 @xmath7 m in the region where the pli photocathode would be placed .
note that this region was about 45 mm in length because of the spherical shape of the pli input window . at the input window of pli
, 1 arcmin resolution corresponds to 200 @xmath7 m .
considering the fact that fwhm is 2.35 @xmath8 in the case of a gaussian distribution , fwhm of 470 @xmath7 m corresponds to 1 arcmin resolution by using a 1-@xmath8 deviation as the measure of the resolution .
thus , we concluded that our laser system could be used as the light source of psf measurements since obtained fwhms of @xmath1120 @xmath7 m were well below 1 arcmin resolution at the input window .
[ fig : psf ] shows examples of psf spot images taken with the above explained laser system ; the left and right panels show spot images in the central and peripheral regions , respectively . from top
to bottom the panels show a spot diagram , the cross section along the major axis ( determined from the elliptical shape of the spot ) , and the cross - section along the minor axis ( taken with the axis perpendicular to the major axis ) . a good resolution of fwhm@xmath250 @xmath7 m was achieved at the central region , though there was some deterioration at the peripheral regions due to an increase in the tails of the distributions . in those spot diagrams ,
radial directions correspond to @xmath945@xmath0 .
it should be noted that the directions of the tail in the peripheral regions tend to turn away from the radial direction .
since the pli should have revolution symmetry along its axis , this trend might suggest the asymmetry in alignment of the parts and/or the other effects .
note also that the obtained resolution includes a small contribution from the employed digital camera ( one pixel corresponds to 13.6 @xmath7 m considering the magnification factor of close - up lens mounted on the camera ) .
a half of the pixel size was quadratically added to the quoted error to conservatively indicate the effect of pixelization in the determination of fwhm .
[ fig : psf_dist ] shows the psf dependence on the photon incident position . [ cols="^,^,^ " , ] the closed circles and squares represent the fwhms along the major and minor axes , respectively . with the adopted electric field configuration ,
the magnification factor was measured to be 0.046 ( 0.045 ) at the center ( edge ) , resulting in the output image diameter of @xmath10 23 mm .
we observed a good resolution corresponds to 23 arcmin in most region of the sensitive area .
however , there is an asymmetry between the left and right side resolutions , due possibly to the difference in the axis of the input , output window and the main parts of the pli .
we have succeeded in producing a 20-in .
diameter photoelectric lens image intensifier tube ( pli ) for the first time .
the pli is designed to be mounted on the focal surface of the ashra optical system .
the pli reduces the image size into solid state imaging devices with some amplification .
it enables drastic pixel cost reduction compared to conventional photomultiplier arrays at the focal surface of telescopes .
hence , the pli is a key technology in the ability of ashra to perform all - sky surveys with a few arcmin resolution .
furthermore , because of its large diameter and fine imaging capability , plis may have many other applications in high - energy and astrophysical experiments .
this work was mostly supported by the coordination of science and technology ( 157 - 20004100 ) , and was supported in part by the grant - in - aid for scientific research ( 16740130 ) of the ministry of education , science , sports and culture in japan . | we have evaluated a 20-in .
photoelectric lens image intensifier tube ( pli ) to be mounted on the spherical focal surface of the ashra light collectors , where ashra stands for all - sky survey high resolution air - shower detector , an unconventional optical collector complex that images air showers produced by very high energy cosmic - ray particles in a 42@xmath0-diameter field of view with a resolution of a few arcminutes .
the pli , the worlds largest image intensifier , has a very large effective photocathode area of 20-in .
diameter and reduces an image size to less than 1-inch diameter using the electric lens effect .
this enables us to use a solid - state imager to take focal surface images in the ashra light collector .
thus , pli is a key technology for the ashra experiment to realize a much lower pixel cost in comparison with other experiments using photomultiplier arrays at the focal surface . in this paper
we present the design and performance of the 20-in .
pli .
first generation image intensifier tube ; large sensitive area ; photodetector ; high energy astrophysics ; ashra experiment |
measurements of the diffuse ultraviolet ( uv ) radiation field have to contend with a number of contaminating sources including atmospheric emission lines and the zodiacal light @xcite .
these foreground sources are particularly important at high galactic latitudes where the galactic contribution to the radiation field is relatively small and at longer wavelengths where the zodiacal light , which follows the solar spectrum , becomes increasingly important .
it has been difficult to disentangle these components , largely because of a lack of relevant observations . ideally , these would be spectroscopic observations with moderate resolutions over a large part of the sky with different sun angles .
however , what we have is thousands of observations from the galaxy evolution explorer ( _ galex _ ) in two bands ( fuv : 1531 and nuv : 2321 ) with observations far from the sun to minimize foreground emission . despite these drawbacks
, we have used the _ galex _
data to derive empirical formulae for the foreground sources .
although our main interest is in better understanding the galactic and extragalactic diffuse radiation , we hope that our results will also prove useful in studies of the earth s atmosphere and of the zodiacal light .
they will certainly prove useful in mission planning for other space - borne instruments such as the ultraviolet imaging telescope @xcite which will observe the sky with large field of view instruments where diffuse radiation limits the observable sky .
the _ galex _ spacecraft was launched in 2003 and has since observed about 75% of the sky in two spectral bands ( fuv : 1531 and nuv : 2321 ) with a spatial resolution of about 5 over a field of view of 0.6 .
the primary mission was described by @xcite and the software and data products by @xcite .
most of the _ galex _
observations were short exposures of about 100 seconds in length ( all - sky imaging survey : ais ) but there were a number of longer observations of 10,000 seconds or more , either to fulfill specific mission objectives or taken as part of the guest investigator ( gi ) program .
a single exposure was limited by the duration of the orbital night ( about 1000 seconds ) and longer observations were broken up into a series of exposures spread over a time period ranging from days to years . _ galex _
observations were subject to severe selection effects due to brightness related constraints from the diffuse radiation integrated over the large field of view . thus the galactic plane and other high intensity regions such as orion or the magellanic clouds
could not be observed in the original mission .
recent observations have covered many of these but they will not help in refining the foreground because of the brightness of the astrophysical emission .
observations were only taken at orbital night between 20:00 and 04:00 ( local time ) and only in directions more than 90 from the sun to minimize airglow and zodiacal light .
hence , we only sampled a limited part of the total phase space of observational parameters .
the diffuse background is the sum of the galactic background , which depends only on the look direction , and the foreground emission airglow and zodiacal light which depends on the time and date of the observation .
the standard data products include a single image of the targeted field for each of the two bands ( the fuv detector failed in may , 2009 following which observations were made only with the nuv detector ) and a merged catalog of point sources from both bands .
because we are trying to derive the foreground emission which is a function of time and date of the observation , we used the spacecraft housekeeping files ( scst files ) which included the total count rate ( tec : total event count ) in each of the two detectors .
the tec was tabulated every second and was tied to the ut ( universal time ) of the observation .
a typical tec is plotted with respect to ut in figure [ fig : tec_counter ] . the total event counter ( tec ) as a function of ut .
the minimum emission was at local ( spacecraft ) midnight . ]
as implied by its name , the tec includes all emission in the field of view including starlight , diffuse background , airglow , and zodiacal light and each element had to be estimated separately .
however , only the airglow would be expected to vary with the time of day and , to anticipate our results , specifically with the time from local midnight . unfortunately , this was not readily available from the _ galex _ data products and we obtained the tles ( two - line elements ) from space-track.org ( ` https://www.space-track.org/ ` ) and then used stk ( ` http://www.agi.com/default.aspx ` ) to calculate the latitude and longitude of the spacecraft ground track at a given ut . from this
we calculated the local spacecraft time and cross - indexed with the tec to obtain the total emission as a function of time . as a result of this rather painful experience
, we would recommend that future missions include the local spacecraft coordinates as part of their standard data products .
there were about 34,000 observations in each of the fuv and nuv bands in the _ galex _
gr6 data release with close to 76,000 scst files and 14,000,000 individual tec measurements in each band .
note that the longer exposures were divided into multiple exposures and hence multiple scst files .
naturally enough , most observations of the airglow from space have been downward looking and have found a number of different atomic and molecular lines ( reviewed by @xcite ) .
there are many fewer observations looking up from low - earth orbit ( leo ) .
the only emission lines observable in the night spectrum are the geocoronal o i lines at 1304 ( 0.013 kr ) and 1356 ( 0.001 kr ) in the fuv band and 2471 ( @xmath0 kr ) in the nuv band @xcite .
the _ galex _
fuv band rejected the 1304 line but with a 10% leak @xcite and these values corresponded to expected levels of about 200 photons @xmath1 s@xmath2 sr@xmath2 @xmath2 in the fuv band and 100 photons
@xmath1 s@xmath2 sr@xmath2 @xmath2 in the nuv band .
as mentioned above , we have adopted an empirical approach in studying the foreground emission . we noted that each observation could be separated into two parts ; a minimum value at orbital midnight and a time - variable part which increased smoothly on either side of orbital midnight ( figure [ fig : tec_counter ] ) .
the only possible source for a signal that varies with the local time is airglow and we extracted this component of the foreground emission by subtracting a baseline calculated from the average of the points within 15 minutes of local midnight , assuming that the observation included this time span .
this baseline is comprised of all other sources in the field , including any residual airglow emission at midnight .
there were approximately 5.8 million independent points in the fuv channel and 6.6 million in the nuv channel with an overlap of about 5 million points .
we gridded the baseline - subtracted data to form a density plot for each band and these are shown in figure [ fig : ag_density ] .
these plots were created by gridding the data into bins of 10 minutes in time and 10 photons @xmath1 s@xmath2 sr@xmath2 @xmath2 in flux .
the plus signs show the peak airglow at a given time and the 1@xmath3 contour is shown as the closed line ; that is , 68% of the data points fall within the two contour lines between the observation limits of 20:00 to 04:00 .
the dark lines are the best fit quadratics to the airglow emission in each band with the equations : @xmath4 @xmath5 where fuv and nuv are the total flux in the respective bands in units of photons @xmath1 s@xmath2 sr@xmath2 @xmath2 and t is the time from midnight in hours . if the observation stretches the entire 8 hours , the maximum for a _ galex _
observation , the airglow contribution would be 25 and 33 photons @xmath1 s@xmath2 sr@xmath2 @xmath2 in the fuv and nuv bands , respectively , with an estimated error of about 20 photons @xmath1 s@xmath2 sr@xmath2 @xmath2 primarily due to the difficulty of measuring the baseline .
note that the time measured is time on the ground track where 8 hours corresponds to about 20 minutes of actual spacecraft time .
there is a strong correlation between the fuv and nuv data ( r = 0.745 ) with nuv = 0.7 fuv , consistent with the origin of both in geocoronal o i lines .
the zodiacal light is due to sunlight scattered by interplanetary dust in the uv and visible and thermal emission from the dust in the infrared , with a black body temperature of about 60 k. it has been observed extensively in the visible from ground - based observations ( reviewed by @xcite ) and in the infrared @xcite from the _ infrared astronomy satellite _ ( _ iras _ ) mission .
it is generally assumed that the spectrum of the zodiacal light follows the solar spectrum ; that is , the color of the zodiacal light is unity .
however , there have been very few observations in the ultraviolet with the most robust being those of @xcite who claimed that the color of the zodiacal light increased with the ecliptic angle . in any case
, the zodiacal light will only contribute in the nuv band .
we further restricted our dataset to those observations which included both fuv and nuv data and where the observations extended at least 15 minutes on either side of orbital midnight such that a baseline could be robustly defined .
there were 9313 such observations and we extracted the baseline value for each .
this baseline value at orbital midnight includes , as mentioned above , starlight , the galactic background , airglow and , in the nuv band only , the zodiacal light .
the _ galex _
pipeline produces a merged catalog for each observation in which the fluxes of the point sources in each band are tabulated from which we could estimate and subtract the contribution due to starlight in each band .
these data the baseline value at midnight from which the starlight has been subtracted form the basis of our further analysis and the nuv data ( including residual airglow ; diffuse astrophysical background ; and zodiacal light ) are plotted as a function of sun angle in figure [ fig : nuv_baseline ] .
there is , of course , considerable scatter reflecting the range in galactic latitude spanned by the observations but a lower envelope to the values can be cleanly drawn with the rise on the right due to locations at low ecliptic latitudes .
zodiacal light does contribute significantly to the nuv channel and is strongly dependent on the ecliptic latitude ( figure [ fig : zod_with_ecliptic ] ) and the helioecliptic longitude ( ecliptic longitude - longitude of the sun ) .
because of operational constraints on _ galex _ observations , all the data were taken at large helioecliptic longitudes ( figure [ fig : obs_with_earth ] ) and , in fact , there was little variation with longitude .
@xcite have tabulated observations of the zodiacal light in the optical and , as a first estimate , we have assumed that the zodiacal light in the uv follows the optical distribution with a spectral correction given by the solar spectrum ( eg .
@xcite ) . the predicted zodiacal light tracks the observed tec well ( figure [ fig : zl_obs_mod ] ) with the formula : tec = 690 + 0.65 zl , where zl is the zodiacal light from the visible observations by @xcite . the baseline is the zero level representing the residual airglow and the diffuse galactic light while the slope represents the color of the zodiacal light with respect to the visible the scattered light in the uv is 65% that in the optical .
unlike @xcite , we find no dependence of the color on the ecliptic latitude .
although a detailed study of the zodiacal dust is beyond the scope of this work , this suggests that the albedo ( reflectivity ) of the interplanetary dust grains in the uv is 65% of the albedo in the visible , in rough agreement with results for interstellar dust @xcite .
nuv tec as a function of ecliptic latitude . ]
observations as a function of helioecliptic longitude and latitude . ] observed nuv tec as a function of predicted zodiacal light . ]
we have subtracted starlight from both bands and zodiacal light from the nuv band and plotted the dependence of these on the sun angle in figure [ fig : ag_baseline ] .
note that the rise in data values on the right side of figure [ fig : nuv_baseline ] is no longer seen , indicating that it is , indeed , due to zodiacal light .
we have empirically defined a lower envelope to the fuv and nuv tec by the equations @xmath6 @xmath7 where sa is the angle from the sun in radians and fuv and nuv are the respective tec values in units of photons @xmath1 s@xmath2 sr@xmath2 @xmath2 .
the variable component must be due to the airglow in the fuv , where zodiacal light will not contribute , and hence is likely to be from the airglow in the nuv also , where it follows the same dependence on sun angle .
however , there is no way to independently separate the astrophysical component from the baseline airglow . if we take recourse to the canonical value of 300 photons @xmath1
s@xmath2 sr@xmath2 @xmath2 for the diffuse background @xcite , we find that the fuv airglow is 220 photons @xmath1
s@xmath2 sr@xmath2 @xmath2 at midnight and the nuv airglow is 350 photons @xmath1
s@xmath2 sr@xmath2 @xmath2 .
we have empirically derived predictions for the foreground emission in the two _ galex _ bands .
there are two components to the airglow : one which is dependent on the local time and is symmetrical around local midnight ; and the other which is dependent on the angle between the target and the sun .
we find that the zodiacal light is proportional to that in the visible with a color of 0.65 .
we are left with an ambiguity in the baseline at the 100 - 200 photons @xmath1 s@xmath2 sr@xmath2 @xmath2 level which will require spectroscopic information to disentangle . | ultraviolet observations from low earth orbit ( leo ) have to deal with a foreground comprised of airglow and zodiacal light which depend on the look direction and on the date and time of the observation .
we have used all - sky observations from the galex spacecraft to find that the airglow may be divided into a baseline dependent on the sun angle and a component dependent only on the time from local midnight .
the zodiacal light is observable only in the near ultraviolet band ( 2321 ) of galex and is proportional to the zodiacal light in the visible but with a color of 0.65 indicating that the dust grains are less reflective in the uv . |
the initial observations by the batse instrument on board the compton @xmath6-ray observatory satellite have evidenced what has later become known as the prompt radiation of grbs . on the basis of their hardness as well as their duration , grbs were initially classified into short and long at a time when their cosmological nature was still being disputed .
the advent of the _ bepposax _ satellite @xcite introduced a novel approach to grbs by introducing joint observations in the x - rays and @xmath6-rays thanks to its instruments : the gamma - ray burst monitor ( @xmath7@xmath8 kev ) , the wide field cameras ( @xmath9@xmath10 kev ) , and the narrow field instruments ( @xmath9-@xmath11 kev ) .
the unexpected and welcome discovery of the existence of a well separate component in the grb soon appeared : the afterglow radiation lasting up to @xmath12@xmath13 s after the emission of the prompt radiation .
beppo - sax clearly indicated the existence of a power law behavior in the late x - ray emission ( lxre ; see fig .
[ bepposax ] ) .
the coming of the _ swift _ satellite , significantly extending the observation in the x - ray band thanks to its x - ray telescope ( xrt band : @xmath14@xmath11 kev ) , has allowed us for the first time to cover the unexplored region between the end of the prompt radiation and the power law late x - ray behavior discovered by _ bepposax _ : in some long grbs a steep decay phase was observed followed by a plateau leading then to a typical lxre power law behavior .
already , @xcite noticed the unexpected result that the lxres of a `` golden sample '' ( gs ) of six long , closeby ( @xmath15 ) , energetic ( @xmath16 erg ) grbs , when measured in the rest - frame of the sources , were showing a common power law behavior ( see fig .
[ scaling ] ) , independently from the isotropic energy @xmath0 coming from the prompt radiation ( see fig [ nesting ] ) .
more surprising was the fact that the plateau phase luminosity and duration before merging in the common lxre power law behavior were clearly functions of the @xmath0 , while the late power law remains independent from the energetic of the prompt radiation .
for this reason , this remarkable scaling law has been used as a standard candle to independently estimate the cosmological redshift of some long grbs by requiring the overlap of their lxre ( see , e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ) , and also to predict , 10 days in advance , the emergence of the typical optical signature of the supernova sn 2013cq , associated with grb 130427a @xcite .
the current analysis is based on the paradigms introduced in @xcite for the spacetime parametrization of the grbs , in @xcite for the interpretation of the structure of the grb prompt emission , and in @xcite for the induced gravitational collapse ( igc ) process , further evolved in @xcite,@xcite , @xcite , and @xcite . in the present case
, the phenomenon points to an igc occurring when a tight binary system composed of a carbon - oxygen core ( co@xmath2 ) undergoes a supernova ( sn ) explosion in the presence of a binary neutron star ( ns ) companion @xcite .
when the igc leads the ns to accrete enough matter and therefore to collapse to a black hole ( bh ) , the overall observed phenomenon is called binary - driven hypernova ( bdhn ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) .
s. ] a crucial further step has been the identification as a bdhn of grb 090423 at the extreme redshift of @xmath17 @xcite . on top of that , the lxre of grb 090423 overlaps perfectly with the ones of the gs ( see fig . [ z8overlap ] ) , extending such a scaling law up to extreme cosmological distances .
this result led to the necessity of checking such an universal behavior of the lxres in bdhne at redshifts larger than @xmath18 ( see the sample list in table [ estable ] ) .
it is clear by now that the afterglow analysis is much more articulated than previously expected and contains new specific signatures .
when theoretically examined within our framework , these new signatures lead to specific information on the astrophysical nature of the progenitor systems @xcite . in the present paper , we start by analysing the signatures contained in the lxres at @xmath19 s , where @xmath20 is the rest - frame time after the initial grb trigger . in particular
, we probe a further improvement for the existence of such an lxre universal behavior of bdhne by the introduction of a collimation correction
. in section 2 we present an `` enlarged sample '' ( es ) of 161 bdhne observed up to the end of 2015 .
in particular , we express for each bdhn : ( 1 ) redshift ; ( 2 ) @xmath0 ; and ( 3 ) the lxre power law properties .
we probe the universality of the lxre power law behavior as well as the absence of correlation with the prompt radiation phase of the grb . in section 3
we introduce the collimation correction for the lxre of bdhne .
this , in turn , will aim to the possible establishment of a new cosmological candle , up to @xmath21 . in section 4
we present the inferences for the understading of the afterglow structure , and , in section 5 , we draw our conclusions
. + @xmath22 ( a ) ( b ) @xmath22 +
we have built a new sample of bdhne , which we name `` enlarged sample '' ( es ) , under the following selection criteria : * measured redshift @xmath23 ; * grb rest - frame duration larger than @xmath9 s ; * isotropic energy @xmath0 larger than @xmath24 erg ; and * presence of associated _
swift_/xrt data lasting at least up to @xmath25 s. we collected @xmath26 sources , which satisfy our criteria , covering @xmath27 years of _ swift_/xrt observations , up to the end of 2015 , see table [ estable ] .
the @xmath0 of each source has been estimated using the measured redshift @xmath23 together with the best - fit parameters of the @xmath6-ray spectrum published in the gcn circular archive .
the majority of the es sources , 102 out of 161 , have @xmath6-ray data provided by _
fermi_/gbm and konus - wind , which , with their typical energy bands being @xmath11@xmath28 kev and @xmath29@xmath30 kev , respectively , lead to a reliable estimate of the @xmath0 , computed in the `` bolometric '' @xmath31@xmath32 kev band @xcite .
the remaining sources of the es have had their @xmath6-ray emission observed by _
swift_/bat only , with the sole exception of one source observed by hete .
the energy bands of these last two detectors , being @xmath33@xmath34 kev and @xmath35@xmath36 kev , respectively , lead to a standard estimate of @xmath0 by extrapolation in the `` bolometric '' @xmath31@xmath32 kev band @xcite .
+ @xmath22 ( a ) ( b ) @xmath22 + @xmath22 + + @xmath22 ( c ) ( d ) @xmath22 + we compare the _ swift_/xrt isotropic luminosity light curve @xmath37 for 161 grbs of the es in the common rest - frame energy range of @xmath38@xmath39 kev .
we initially convert the observed _ swift_/xrt flux @xmath40 as if it had been observed in the @xmath38@xmath39 kev rest - frame energy range . in the detector frame ,
the @xmath38@xmath39 kev rest - frame energy range becomes @xmath41\,$]@xmath42 $ ] kev , where @xmath23 is the redshift of the grb .
we assume a simple power law function as the best fit for the spectral energy distribution of the _ swift_/xrt data : @xmath43 therefore , we can compute the flux light curve in the @xmath38@xmath39 kev rest - frame energy range , @xmath44 , multiplying the observed one , @xmath40 , by the k - correction factor : @xmath45 then , to compute the isotropic x - ray luminosity @xmath46 , we have to multiply @xmath44 by the spherical surface having the luminosity distance as radius @xmath47 where we assume a standard cosmological @xmath48cdm model with @xmath49 and @xmath50 . finally , we convert the observed times into rest - frame times @xmath20 : @xmath51 we then fit the whole isotropic luminosity light - curve late phase with a decaying power law function defined as follows : @xmath52 where @xmath53 , the power law index , is a positive number , and @xmath54 is the luminosity at an arbitrary time @xmath55 after the grb trigger in the rest - frame of the source .
all the power laws are shown in fig .
[ descriptive]b .
[ descriptive2]a shows the distribution of the @xmath53 indexes within the es .
such a distribution follows a gaussian behavior with a mean value of @xmath56 and a standard deviation of @xmath57 .
the lxre luminosity light curves of the es in the @xmath14@xmath11 kev rest - frame energy band are plotted in fig .
[ descriptive]a , compared to the curves of the gs .
[ descriptive]a shows that the power laws within the es span around two orders of magnitude in luminosity .
the spread of the lxre light curves in the es is better displayed by fig .
[ descriptive2]b which shows the distribution within the es of the lxre integrated energies @xmath58 defined as : @xmath59 we choose to represent the spread of the lxre luminosity light curves with the late integrated energy @xmath58 at late times ( @xmath25@xmath13 s ) instead of the luminosity @xmath46 at a particular time for two reasons : ( 1 ) there is no evidence for a particular time in which to compute @xmath46 and ( 2 ) we want to have a measurement of the spread as much as possible independent from the slopes , whose dispersion causes the mixing of part of the light curves over time ( see fig .
[ descriptive]b ) .
the integration time interval @xmath25@xmath13 reasonably contains most of the data in the late power law behavior .
in fact , the lower limit @xmath25 s is basically the average of the initial time for our linear fits on the data ( more precisely @xmath60 s ) , while the upper limit @xmath61 has been chosen because only @xmath62 of the es have x - ray data over such rest - frame time . the solid red line in fig .
[ descriptive2]b represents the gaussian function that best fits the late integrated energies @xmath58 in logarithmic scale .
its mean value is @xmath63 , while its standard deviation is @xmath64 .
the lxre power - law spread , given roughly by @xmath65 , is larger in respect to the previous work of @xcite , which results as @xmath66 .
this is clearly due to the significant growth of the number of bdhne composing the es ( 161 ) in respect to the ones of the gs ( 6 ) .
furthermore , fig .
[ descriptive2]c - d show the scatter plots of the values of @xmath53 and @xmath58 , respectively , versus the values of the @xmath0 for all the sources of the es . in both cases the correlation factors , @xmath67 and @xmath68
respectively , are low , confirming that there is no evidence for a correlation between the lxre power law behavior and the isotropic energy emitted by the source during the prompt radiation .
these results address a different aspect than the ones by @xcite .
there , the authors , after correctly noticing the difficulties of the traditional afterglow model @xcite , attempt to find a model - independent correlation between the x - ray light curve observed in both short and long grbs with their prompt emission . in their work ,
@xcite have considered the integrated x - ray emission over the entire light curve observed by xrt , following @xmath69 s after the grb trigger both for short and long grbs .
such an emission is clearly dominated by the contribution at @xmath70 s , where a dependence from the @xmath0 is self - evident from the above fig .
[ scaling ] and fig .
[ nesting ] .
our approach instead solely applies to the bdhne : ( 1 ) long grbs , and ( 2 ) @xmath16 erg .
this restricts the possible sources in the @xcite sample to @xmath71 grbs : in the present article , we consider a larger sample of @xmath26 bdhne . moreover , ( 3 ) our temporal window starts at @xmath19 s. under these three conditions , our result of the universal lxre behavior has been found .
+ @xmath22 ( a ) ( b ) @xmath22 + we here propose to reduce the spread of the lxre power laws within the es by introducing a collimation effect in the emission process .
in fact , if the emission is not isotropic , fig .
[ descriptive]a - b should actually show overestimations of the intrinsic lxre luminosities . by introducing a collimation effect , namely assuming that the lxres are not emitted isotropically but within a double - cone region having half - opening angle @xmath72
, we can convert the isotropic @xmath73 to the intrinsic lxre luminosity @xmath74 as : @xmath75 from eq .
[ lumintrinsec ] , an angle @xmath72 can be inferred for each source of the es if an intrinsec universal lxre light curve @xmath74 is assumed .
for example , assuming an intrinsic standard luminosity @xmath76 , at an arbitrary time @xmath55 , eq .
[ lumintrinsec ] becomes @xmath77 , which , in principle , could be used to infer @xmath72 for each source . on the other hand , for the same reasons expressed in the previous section , we choose to estimate the angles @xmath72 using the lxre integrated energy @xmath58 instead of the @xmath46 at a particular time .
therefore , we simply integrate eq .
[ lumintrinsec ] in the rest - frame time interval @xmath32@xmath13 s : @xmath78 obtaining , consequentially @xmath79 by assuming a universal @xmath80 for all bdhne , it is possible to infer @xmath72 for each source of the es .
we assume grb 050525a , having the lowest @xmath58 within the es , as our unique `` isotropic '' source , namely , in which we can impose @xmath81 , which automatically gives @xmath82 , which means that the lxre luminosity is emitted over all the isotropic sphere . on the top of having the weakest lxre over 11 years of _ swift_/xrt observations , grb 050525a : _ ( a )
_ has been observed by konus - wind in the @xmath6-rays @xcite , then its @xmath0 estimate is reliable ; _ ( b ) _ has a reliable late x - ray slope given by a complete _ swift_/xrt light curve ( showing a late power law behavior from @xmath83 to @xmath84 s in the rest frame ) ; _ ( c ) _ has an associated supernova @xcite .
an instrumental selection effect can not affect this choice since _ swift_/xrt can easily detect and follow x - ray emissions weaker than that of grb 050525a .
some examples are the two x - ray luminosity light curves shown of grb 060218 and grb 101219b in fig .
[ xrf]-a . furthermore , in the case of a future observation of a bdhn showing a @xmath58 weaker than the one of grb 050525a , a renormalization of the angle distribution of fig .
[ angles]a down to lower angle values with respect to the new minimum one would be required , leaving the overall angle distribution unaltered .
all of these facts make grb 050525a a robust `` isotropic '' bdhn candidate . with @xmath85
now fixed , a half - opening angle @xmath72 is inferred for each source of the es using eq .
the values of @xmath72 are listed in table [ estable ] .
[ angles]a shows the probability distribution of the half - opening angle @xmath72 within the es .
the blue solid line represents a logarithmic normal distribution , which best fits the data .
this distribution has a mode of @xmath86 , a mean of @xmath87 , a median of @xmath88 , and a standard deviation of @xmath89 .
moreover , it is possible to verify that , by correcting the @xmath37 light curve of each es source for its corresponding @xmath72 , an overlap of the lxre luminosity light curves as good as the one seen in the gs by @xcite shown in fig .
[ scaling ] is obtained .
since the lxre follows a power law behavior , we can quantify the tightness of the lxres overlap looking at the correlation coefficient @xmath90 between all the luminosity light - curve data points of the es sources in log - log scale .
considering the data points of the lxre power laws within the @xmath32@xmath13 s time interval ( where we have defined @xmath58 ) , we obtain @xmath91 for the gs , @xmath92 for the es before the collimation correction , and @xmath93 after the correction .
therefore , the collimation correction not only reduces the spread of the lxres within the es , but makes the lxres overlap even tighter than the one previously found in the gs . finally , in order to test the robustness of our results , we do the same analysis excluding , by the es , the sources seen only by _
swift_/bat or hete and not by _ fermi_/gbm or konus - wind , under the hypothesis that their @xmath0 estimates are unreliable .
the results obtained using this new sample called es2 , namely the typical value and the dispersion of @xmath53 , @xmath58 , and @xmath72 are summarized in table [ tabeses2 ] .
there is no significant difference between the results obtained from the two samples .
therefore , we conclude that a possible wrong estimate of @xmath0 for the sources observed by only _ swift_/bat or hete and not by _ fermi_/gbm or konus - wind does not bias our results .
.summary of the results of this work obtained by the complete es sample in comparison with the ones arising using the sample es2 , namely the es deprived by the sources seen by _ swift_/bat or hete only . [ cols="<,^,^",options="header " , ] [ tabeses2 ]
in the last 25 years we have seen in the grb community a dominance of the fireball model , which sees the grb as a single astrophysical system , the `` collapsar '' , originating from an ultra - relativistic jetted emission described by the synchrotron / self - synchrotron compton ( ssc ) and traditional afterglow models .
such methods have been systematically adopted to different types of grbs like , for example , the short , hard grb 090510 @xcite , the high energetic long grb 130427a @xcite , the low energetic short grb 051221a @xcite , and the low energetic long grb 060218 @xcite , independently from the nature of their progenitors . in the recent four years
, substantial differences among seven distinct kinds of grbs have been indicated , presenting different spectral and photometrical properties on different time scales @xcite .
the discovery of several long grbs showing multiple components and evidencing the presence of a precise sequence of different astrophysical processes during the grb phenomenon ( e.g. * ? ? ? * ; * ? ? ?
* ) , led to the introduction of a novel paradigm expliciting the role of binary sources as progenitors of the long grb - sn connection .
this has led to the formulation of the igc paradigm @xcite . within the igc paradigm ,
a tight binary system composed of a carbon - oxygen core ( co@xmath2 ) undergoing a supernova ( sn ) explosion in the presence of a binary ns companion has been suggested as the progenitor for long gamma - ray bursts .
different scenarios occur depending on the distance between the co@xmath2 and the ns binary companion @xcite .
correspondingly two different sub - classes of long bursts have been shown to exist ( for details , see * ? ? ?
* ; * ? ? ?
a first long burst sub - class occurs when the co@xmath2ns binary separation @xmath94 is so large ( typically @xmath95 cm , see , e.g. , @xcite that the accretion of the sn ejecta onto the ns is not sufficient to have the ns reach its critical mass , @xmath96 , for gravitational collapse to a bh to occur .
the hypercritical accretion of the sn ejecta onto the ns binary companion occurs in this case at rates below @xmath97 solar masses per second and is characterized by a large associated neutrino emission @xcite .
we refer to such systems as x - ray flashes ( xrfs ) .
a second long burst sub - class occurs when the co@xmath2ns binary is more tightly bound ( @xmath98 cm , see , e.g. , @xcite
. the larger accretion rate of the sn ejecta , e.g. , @xmath97@xmath99 solar masses per second , leads the companion ns to easily reach its critical mass @xmath96 @xcite , leading to the formation of a bh .
we refer to such systems as binary - driven hypernovae . a main observational feature , which allows us to differentiate bdhne from xrfs is the isotropic @xmath6-ray energy @xmath0 being larger than @xmath24 erg .
such a separation energy value is intimately linked to the binary separation @xmath94 of the binary progenitor and the consequent birth or not of the bh ( for details , see * ? ? ?
+ thanks to the xrt instrument on board the _ swift _ satellite , we can compare and contrast the x - ray afterglow emissions of bdhne and xrfs ( see fig .
[ xrf ] ) .
the typical x - ray afterglows of xrfs can be divided into two main parts : an initial bump with rapid decay , followed by an emerging slight decaying power law ( see fig .
[ xrf ] ) .
the typical x - ray afterglow light curve of bdhne can be divided into three different parts : ( 1 ) an initial spike followed by an early steep power law decay ; ( 2 ) a plateau phase ; and ( 3 ) a late power law decay , the lxre , which is the presented in this work @xcite .
the treatment of the first parts of the x - ray afterglow of bdhne , namely the spike , the initial steep decay , and the plateau phase , indeed fundamental within the bdhn picture , is beyond the scope of this article and it will be extensively shown in forthcoming works ( * ? ? ?
* ; * ? ? ?
* both in preparation ) .
the universalities of the lxre outlined in this article are then explained within the igc paradigm , originating from the interaction of the grb with the sn ejecta .
the constancy of the late power law luminosity in the rest frame is now explained in terms of the constancy in mass of the sn ejecta , which is standard in a bdhn @xcite .
it is appropriate to point out that no achromatic `` jet - break '' effect has been observed in any of the 161 sources of our es .
we recall that the achromatic `` jet - break '' effect is a consequence of relativistic jet pictures ( lorentz factor @xmath100@xmath101 ) , in which a change of slope is expected in the late x - ray light curve , which clearly does not apply in the case of the bdhn following the igc model . in this scenario ,
a velocity of expansion @xmath102 ( lorentz factor @xmath103 ) is found , indicating that the collimation of the sn ejecta originates in a mildly relativistic regime .
this can not be related to the ultra - relativistic jet emission recalled above , considered in the early work of @xcite and continued all the way to the more recent results presented by @xcite .
these authors attempted to explain all grbs as originating from a single object with an intrinsic energy approximately of @xmath104 erg @xcite or @xmath105 erg @xcite : the different energetics and structures of all the grbs were intended to be explained by the beaming effect with different ultra - relativistic lorentz factors @xmath100@xmath101 .
indeed , it is by now clear ( see * ? ? ? * ; * ? ? ?
* ) that at least seven different classes of grbs exist , each with different progenitors , different energies , and different spectra . in no way these distinct classes can be explained by a single common progenitor , using simply relativistic beaming effects .
in this work , we give new statistical evidence for the existence of a universal behavior for the lxres of bdhne , introducing the presence of a collimation effect in such emission , and presenting the common lxre energy @xmath85 as a standard candle .
we build an `` enlarged sample '' ( es ) of 161 bdhne , and focus on their lxres and then we introduce a collimation effect .
these analyses lead us to the following results .
* 1 ) * we find for the es an increased variability in the decaying lxre power law behavior in respect to the result previously deduced by @xcite .
the typical slope of the power law characterizing the lxre is @xmath106 ( gs : @xmath107 ) , while the late - time integrated luminosity between @xmath32@xmath13 seconds in the rest frame is @xmath108 ( gs : @xmath109 ) . *
2 ) * the introduction of a collimation in the lxre recovers a universal behavior . assuming a double - cone shape for the lxre region , we obtain a distribution of half - opening angles peaking at @xmath3 , with a mean value of @xmath4 , and a standard deviation of @xmath5 , see fig .
[ angles]a . *
3 ) * the application of the collimation effect to the lxres of the es indeed reduces the scattering of the power law behavior found under the common assumption of isotropy ; see fig .
[ descriptive]a - b .
the power law scattering of the lxres , after being corrected by the collimation factor , results in being even lower than the one found in the gs ; see fig .
[ angles]b .
the fact that these extreme conditions neither were conceived nor are explained within the traditional ultra - relativistic jetted ssc model , in view also of the clear success of the igc paradigm in explaining the above features , comes as a clear support to a model for grbs strongly influenced by the binary nature of their progenitors , involving a definite succession of selected astrophysical processes for a complete description of the bdhne .
it is remarkable that the universal behavior occurs in the rest - frame time interval @xmath32@xmath13 s , which precisely corresponds to the temporal window of the early observations of beppo - sax at the time of the afterglow discovery ( see fig [ bepposax ] ) .
we thank both the editor and the referee for the fruitful correspondence that improved the presentation of our results . this work made use of data supplied by the uk swift science data center at the university of leicester .
j. a. r. acknowledges the support by the international cooperation program capes - icranet financed by capes - brazilian federal agency for support and evaluation of graduate education within the ministry of education of brazil .
m. k. and y. a. acknowledge the support given by the international relativistic astrophysics erasmus mundus joint doctorate program under the grants 20131471 and 20140707 , respectively , from eacea of the european commission .
m. m. acknowledges the partial support of the project no .
3101/gf4 ipc-11/2015 , and the target program of the ministry of education and science of the republic of kazakhstan .
+ @xmath110 grb @xmath110 & @xmath111 z @xmath111 & @xmath111 @xmath53 @xmath111 & @xmath112 @xmath113 @xmath112 & @xmath112 @xmath72 @xmath114 ( @xmath115 ) @xmath112 & @xmath112 @xmath116@xmath117 & @xmath110 instrument@xmath118 @xmath110 & @xmath119 gcn @xmath119 + + grb & z & @xmath53 & @xmath113 & @xmath72 ( @xmath115 ) & @xmath116 & instrument@xmath118 & gcn + 050315a@xmath120 & 1.95 & 0.838 & 15.6 & 10.1 & 8.32 & swift & 3099 + 050318a@xmath120 & 1.44 & 1.74 & 0.442 & 63.3 & 3.70 & swift & 3134 + 050319a@xmath120 & 3.24 & 1.27 & 12.3 & 11.4 & 20.2 & swift & 3119 + 050401a@xmath120 & 2.9 & 1.59 & 3.82 & 20.6 & 92.0 & kw & 3179 + 050408a@xmath120 & 1.24 & 1.14 & 2.19 & 27.2 & 10.8 & hete & 3188 + 050505a@xmath120 & 4.27 & 1.41 & 8.66 & 13.6 & 44.6 & swift & 3364 + 050525a@xmath120 & 0.606 & 1.47 & 0.243 & 90.0 & 2.31 & kw & 3474 + 050730a@xmath120 & 3.97 & 2.42 & 1.74 & 30.7 & 42.8 & swift & 3715 + 050802a@xmath120 & 1.71 & 1.55 & 1.24 & 36.5 & 7.46 & swift & 3737 + 050814a@xmath120 & 5.3 & 2.23 & 2.90 & 23.6 & 26.8 & swift & 3803 + 050820a@xmath120 & 2.61 & 1.23 & 22.1 & 8.51 & 39.0 & kw & 3852 + 050922c@xmath120 & 2.2 & 1.57 & 0.974 & 41.4 & 19.8 & kw & 4030 + 051109a@xmath120 & 2.35 & 1.20 & 9.09 & 13.3 & 18.3 & kw & 4238 + 060108a@xmath120 & 2.03 & 0.830 & 3.40 & 21.8 & 1.50 & swift & 4445 + 060115a@xmath120 & 3.53 & 1.58 & 2.23 & 27.0 & 19.0 & swift & 4518 + 060124a@xmath120 & 2.296 & 1.30 & 20.7 & 8.79 & 26.0 & kw & 4599 + 060202a@xmath120 & 0.783 & 1.03 & 4.57 & 18.8 & 1.90 & swift & 4635 + 060206a@xmath120 & 4.05 & 1.44 & 28.8 & 7.45 & 3.90 & swift & 4697 + 060210a@xmath120 & 3.91 & 1.75 & 25.9 & 7.86 & 120 . & swift & 4734 + 060418a@xmath120 & 1.49 & 1.54 & 0.614 & 52.9 & 12.0 & kw & 4989 + 060502a@xmath120 & 1.51 & 1.26 & 2.96 & 23.4 & 11.0 & swift & 5053 + 060510b@xmath120 & 4.9 & 1.54 & 3.39 & 21.8 & 66.0 & swift & 5107 + 060512a@xmath120 & 2.1 & 1.24 & 0.356 & 71.5 & 2.38 & swift & 5124 + 060526a@xmath120 & 3.21 & 2.27 & 1.36 & 34.7 & 5.40 & swift & 5174 + 060605a@xmath120 & 3.8 & 2.00 & 0.524 & 57.6 & 10.1 & swift & 5231 + 060607a@xmath120 & 3.082 & 3.04 & 0.481 & 60.4 & 34.0 & swift & 5242 + 060707a@xmath120 & 3.43 & 1.18 & 11.4 & 11.8 & 5.30 & swift & 5289 + 060708a@xmath120 & 1.92 & 1.21 & 1.22 & 36.8 & 2.20 & swift & 5295 + 060714a@xmath120 & 2.71 & 1.47 & 1.93 & 29.0 & 19.0 & swift & 5334 + 060729a@xmath120 & 0.54 & 1.31 & 3.36 & 21.9 & 1.60 & swift & 5370 + 060814a@xmath120 & 0.84 & 1.16 & 1.69 & 31.1 & 6.30 & kw & 5460 + 060906a@xmath120 & 3.685 & 1.47 & 0.482 & 60.3 & 25.9 & swift & 5534 + 061007a@xmath120 & 1.261 & 1.68 & 0.477 & 60.6 & 88.0 & kw & 5722 + 061121a@xmath120 & 1.314 & 1.50 & 3.75 & 20.7 & 27.0 & kw & 5837 + 061126a@xmath120 & 1.159 & 1.30 & 3.06 & 23.0 & 8.10 & swift & 5860 + 061222a@xmath120 & 2.088 & 1.52 & 18.7 & 9.25 & 33.0 & kw & 5984 + 070110a@xmath120 & 2.35 & 1.10 & 3.20 & 22.5 & 11.9 & swift & 6007 + 070306a@xmath120 & 1.5 & 1.58 & 8.22 & 14.0 & 15.5 & swift & 6173 + 070318a@xmath120 & 0.84 & 1.42 & 1.97 & 28.8 & 4.37 & swift & 6212 + 070508a@xmath120 & 0.82 & 1.60 & 1.08 & 39.1 & 7.57 & kw & 6403 + 070529a@xmath120 & 2.5 & 1.22 & 1.79 & 30.2 & 31.9 & swift & 6468 + 070802a@xmath120 & 2.45 & 1.41 & 0.633 & 52.0 & 1.64 & swift & 6699 + 071003a@xmath120 & 1.6 & 1.75 & 1.61 & 31.9 & 35.8 & kw & 6849 + 080210a@xmath120 & 2.64 & 1.38 & 1.26 & 36.2 & 11.8 & swift & 7289 + 080310a@xmath120 & 2.43 & 1.53 & 1.34 & 35.1 & 20.0 & swift & 7402 + 080319b@xmath120 & 0.94 & 1.59 & 1.99 & 28.6 & 122 .
& kw & 7482 + 080319c@xmath120 & 1.95 & 1.72 & 6.40 & 15.8 & 14.6 & kw & 7487 + 080605a@xmath120 & 1.64 & 1.59 & 1.39 & 34.5 & 22.1 & kw & 7854 + 080607a@xmath120 & 3.04 & 1.57 & 1.77 & 30.4 & 187 . & kw & 7862 + 080721a@xmath120 & 2.6 & 1.62 & 8.93 & 13.4 & 132 .
& kw & 7995 + 080804a@xmath120 & 2.2 & 1.68 & 1.08 & 39.3 & 56.9 & swift & 8067 + 080805a@xmath120 & 1.51 & 1.07 & 1.21 & 36.9 & 9.96 & swift & 8068 + 080810a@xmath120 & 3.35 & 1.57 & 1.77 & 30.4 & 76.1 & kw+swift & 8101 + 080905b@xmath120 & 2.37 & 1.51 & 4.67 & 18.6 & 9.85 & fermi & 8205 + 080916c@xmath120 & 4.35 & 1.29 & 12.1 & 11.5 & 242 .
& fermi & 8278 + 080928a@xmath120 & 1.69 & 1.69 & 0.564 & 55.3 & 4.93 & fermi & 8316 + 081008a@xmath120 & 1.97 & 1.80 & 0.596 & 53.7 & 9.34 & swift & 8351 + 081028a@xmath120 & 3.04 & 1.66 & 2.62 & 24.9 & 12.0 & swift & 8428 + 081109a@xmath120 & 0.98 & 1.32 & 1.07 & 39.4 & 1.87 &
fermi & 8505 + 081121a@xmath120 & 2.51 & 1.51 & 5.77 & 16.7 & 24.9 & fermi & 8546 + 081203a@xmath120 & 2.1 & 2.00 & 0.446 & 63.0 & 35.1 & swift & 8595 + 081221a@xmath120 & 2.26 & 0.849 & 24.5 & 8.09 & 33.8 & fermi & 8704 + 081222a@xmath120 & 2.77 & 1.40 & 4.28 & 19.4 & 25.9 & fermi & 8715 + 090102a@xmath120 & 1.55 & 1.35 & 1.58 & 32.2 & 19.8 & kw & 8776 + 090313a@xmath120 & 3.38 & 2.72 & 10.4 & 12.4 & 13.1 & swift & 8986 + 090328a@xmath120 & 0.736 & 1.84 & 1.42 & 34.0 & 19.5 & fermi & 9056 + 090418a@xmath120 & 1.61 & 1.68 & 0.967 & 41.5 & 15.0 & kw & 9171 + 090423a@xmath120 & 8.2 & 1.57 & 1.29 & 35.7 & 10.9 & fermi & 9229 + 090424a@xmath120 & 0.544 & 1.18 & 1.76 & 30.5 & 4.51 & fermi & 9230 + 090516a@xmath120 & 3.9 & 1.40 & 4.52 & 18.9 & 65.0 & fermi & 9413 + 090618a@xmath120 & 0.54 & 1.49 & 1.44 & 33.8 & 25.4 & fermi & 9535 + 090715b@xmath120 & 3 . & 1.18 & 4.08 & 19.9 & 22.1 & kw & 9679 + 090809a@xmath120 & 2.74 & 1.57 & 0.727 & 48.3 & 4.98 & swift & 9756 + 090812a@xmath120 & 2.45 & 1.23 & 5.39 & 17.3 & 41.7 & kw+swift & 9821 + 090902b@xmath120 & 1.82 & 1.22 & 6.15 & 16.2 & 403 . & fermi & 9866 + 090926a@xmath120 & 2.11 & 1.50 & 5.29 & 17.4 & 217 . &
fermi & 9933 + 091003a@xmath120 & 0.9 & 1.36 & 1.32 & 35.3 & 10.9 & fermi & 9983 + 091020a@xmath120 & 1.71 & 1.30 & 2.19 & 27.3 & 7.63 & fermi & 10095 + 091029a@xmath120 & 2.75 & 1.60 & 5.81 & 16.6 & 7.52 & swift & 10103 + 091127a@xmath120 & 0.49 & 1.32 & 1.36 & 34.9 & 1.79 & fermi & 10204 + 091208b@xmath120 & 1.06 & 1.09 & 1.15 & 38.0 & 2.03 & fermi & 10266 + 100302a@xmath120 & 4.81 & 0.812 & 5.67 & 16.8 & 5.24 & swift & 10462 + 100425a@xmath120 & 1.75 & 1.20 & 0.767 & 46.9 & 2.84 & swift & 10685 + 100513a@xmath120 & 4.77 & 1.44 & 3.74 & 20.8 & 25.7 & swift & 10753 + 100621a@xmath120 & 0.542 & 1.12 & 8.52 & 13.7 & 6.93 & kw & 10882 + 100814a@xmath120 & 1.44 & 1.64 & 14.3 & 10.6 & 15.8 & fermi & 11099 + 100901a@xmath120 & 1.41 & 1.36 & 7.89 & 14.3 & 8.09 & swift & 11169 + 100906a@xmath120 & 1.73 & 1.87 & 1.32 & 35.3 & 22.5 &
fermi & 11248 + 110128a@xmath120 & 2.339 & 1.16 & 0.620 & 52.6 & 3.60 & fermi & 11628 + 110205a@xmath120 & 2.22 & 1.59 & 0.913 & 42.8 & 41.0 & kw & 11659 + 110213a@xmath120 & 1.46 & 1.81 & 2.33 & 26.4 & 7.00 & fermi & 11727 + 110422a@xmath120 & 1.77 & 1.24 & 6.03 & 16.3 & 62.0 & kw & 11971 + 110503a@xmath120 & 1.613 & 1.36 & 2.45 & 25.7 & 19.0 & kw & 12008 + 110715a@xmath120 & 0.82 & 1.69 & 7.33 & 14.8 & 5.60 & kw & 12166 + 110731a@xmath120 & 2.83 & 1.22 & 6.31 & 16.0 & 42.0 & fermi & 12221 + 110808a@xmath120 & 1.348 & 1.33 & 0.588 & 54.1 & 6.10 & swift & 12262 + 110918a@xmath120 & 0.982 & 1.64 & 19.5 & 9.07 & 190 . & kw & 12362 + 111008a@xmath120 & 4.9898 & 1.66 & 8.76 & 13.5 & 82.0 & kw & 12433 + 111123a@xmath120 & 3.1516 & 1.55 & 2.82 & 24.0 & 70.0 & swift & 12598 + 111209a@xmath120 & 0.67 & 1.54 & 1.45 & 33.6 & 67.0 & kw & 12663 + 111228a@xmath120 & 0.716 & 1.23 & 1.24 & 36.4 & 4.10 & fermi & 12744 + 120119a@xmath120 & 1.73 & 1.28 & 1.68 & 31.2 & 37.5 & fermi & 12874 + 120326a@xmath120 & 1.8 & 1.87 & 9.97 & 12.7 & 3.70 & fermi & 13145 + 120327a@xmath120 & 2.81 & 1.58 & 1.33 & 35.1 & 36.5 & swift & 13137 + 120711a@xmath120 & 1.41 & 1.58 & 18.1 & 9.40 & 166 . & fermi & 13437 + 120712a@xmath120 & 4.17 & 1.37 & 2.90 & 23.7 & 18.3 & fermi & 13469 + 120909a@xmath120 & 3.93 & 1.30 & 11.8 & 11.7 & 73.2 & fermi & 13737 + 120922a@xmath120 & 3.1 & 1.19 & 5.41 & 17.3 & 32.0 & fermi & 13809 + 121024a@xmath120 & 2.3 & 1.44 & 1.18 & 37.4 & 10.7 & swift & 13899 + 121027a@xmath120 & 1.77 & 1.28 & 15.8 & 10.1 & 6.55 & swift & 13910 + 121128a@xmath120 & 2.2 & 1.48 & 1.85 & 29.7 & 13.2 &
fermi & 14012 + 121217a@xmath120 & 3 & 1.26 & 30.6 & 7.23 & 24.2 & fermi & 14094 + 130418a@xmath120 & 1.218 & 1.52 & 0.265 & 85.4 & 9.90 & kw & 14417 + 130420a@xmath120 & 1.297 & 1.25 & 1.32 & 35.4 & 7.74 & fermi & 14429 + 130427a@xmath120 & 0.338 & 1.26 & 4.17 & 19.7 & 105 . & fermi & 14473 + 130427b@xmath120 & 2.78 & 1.85 & 0.275 & 83.4 & 5.04 & swift & 14469 + 130505a@xmath120 & 2.27 & 1.50 & 16.0 & 10.0 & 347 . & kw & 14575 + 130514a@xmath120 & 3.6 & 1.56 & 5.16 & 17.7 & 52.4 & kw+swift & 14702 + 130528a@xmath120 & 1.25 & 1.02 & 2.03 & 28.3 & 4.40 & fermi & 14729 + 130606a@xmath120 & 5.91 & 1.91 & 1.24 & 36.4 & 28.3 & kw & 14808 + 130610a@xmath120 & 2.092 & 1.47 & 0.472 & 61.0 & 6.99 & fermi & 14858 + 130701a@xmath120 & 1.155 & 1.20 & 0.639 & 51.7 & 2.60 & kw & 14958 + 130907a@xmath120 & 1.238 & 1.68 & 7.02 & 15.1 & 304 . & kw & 15203 + 130925a@xmath120 & 0.347 & 1.32 & 2.85 & 23.8 & 18.4 & fermi & 15255 + 131030a@xmath120 & 1.293 & 1.27 & 2.97 & 23.4 & 173 . & kw & 15413 + 131105a@xmath120 & 1.686 & 1.24 & 1.75 & 30.6 & 34.7 & fermi & 15455 + 131108a@xmath120 & 2.4 & 1.72 & 0.932 & 42.3 & 73.0 & fermi & 15477 + 131117a@xmath120 & 4.042 & 1.31 & 0.816 & 45.4 & 1.02 & swift & 15499 + 131231a@xmath120 & 0.642 & 1.44 & 3.39 & 21.8 & 22.2 & fermi & 15644 + 140206a@xmath120 & 2.74 & 1.43 & 13.5 & 10.9 & 35.9 & fermi & 15796 + 140213a@xmath120 & 1.2076 & 2.21 & 16.2 & 9.94 & 9.93 & fermi & 15833 + 140226a@xmath120 & 1.98 & 1.58 & 0.762 & 47.1 & 5.80 & kw & 15889 + 140304a@xmath120 & 5.283 & 1.44 & 7.14 & 15.0 & 13.7 & fermi & 15923 + 140311a@xmath120 & 4.954 & 1.49 & 1.18 & 37.5 & 11.6 & swift & 15962 + 140419a@xmath120 & 3.956 & 1.84 & 9.32 & 13.1 & 186 . & kw & 16134 + 140423a@xmath120 & 3.26 & 1.31 & 4.11 & 19.8 & 65.3 & fermi & 16152 + 140506a@xmath120 & 0.889 & 0.924 & 2.78 & 24.2 & 1.12 & fermi & 16220 + 140508a@xmath120 & 1.027 & 1.40 & 5.46 & 17.2 & 23.9 & fermi & 16224 + 140509a@xmath120 & 2.4 & 1.46 & 0.402 & 66.7 & 9.14 & swift & 16240 + 140512a@xmath120 & 0.725 & 1.62 & 2.22 & 27.1 & 7.76 & fermi & 16262 + 140614a@xmath120 & 4.233 & 1.24 & 2.00 & 28.5 & 7.30 & swift & 16402 + 140620a@xmath120 & 2.04 & 1.43 & 5.16 & 17.7 & 6.22 & fermi & 16426 + 140629a@xmath120 & 2.275 & 1.48 & 0.868 & 44.0 & 6.15 & kw & 16495 + 140703a@xmath120 & 3.14 & 2.21 & 1.82 & 30.0 & 1.72 & fermi & 16512 + 140907a@xmath120 & 1.21 & 0.888 & 2.68 & 24.6 & 2.29 & fermi & 16798 + 141026a@xmath120 & 3.35 & 2.24 & 34.8 & 6.78 & 7.17 & swift & 16960 + 141109a@xmath120 & 2.993 & 1.31 & 7.85 & 14.3 & 5.05 & kw & 17055 + 141121a@xmath120 & 1.47 & 1.67 & 6.56 & 15.6 & 14.6 & kw & 17108 + 141221a@xmath120 & 1.452 & 1.18 & 0.970 & 41.5 & 1.94 & fermi & 17216 + 150206a@xmath120 & 2.087 & 1.28 & 10.1 & 12.6 & 151 . & kw & 17427 + 150314a@xmath120 & 1.758 & 1.57 & 3.08 & 22.9 & 98.1 & fermi & 17579 + 150403a@xmath120 & 2.06 & 1.61 & 26.7 & 7.74 & 91.0 & fermi & 17674 + 150514a@xmath120 & 0.807 & 1.35 & 0.378 & 69.1 & 1.14 & fermi & 17819 + 150821a@xmath120 & 0.755 & 1.20 & 0.884 & 43.5 & 15.1 & fermi & 18190 + 150910a@xmath120 & 1.36 & 1.50 & 0.415 & 65.6 & 22.3 & swift & 18268 + 151021a@xmath120 & 2.33 & 1.47 & 2.69 & 24.6 & 211 . &
kw & 18433 + 151027a@xmath120 & 0.81 & 1.69 & 2.04 & 28.3 & 3.86 & fermi & 18492 + 151027b@xmath120 & 4.063 & 1.75 & 4.19 & 19.6 & 19.1 & swift & 18514 + 151111a@xmath120 & 3.5 & 1.29 & 0.871 & 43.9 & 5.13 & fermi & 18582 + 151112a@xmath120 & 4.1 & 1.54 & 3.31 & 22.1 & 12.4 & swift & 18593 + 151215a@xmath120 & 2.59 & 1.07 & 1.34 & 35.1 & 1.97 & swift & 18699 + | it has previously been discovered that there is a universal power law behavior exhibited by the late x - ray emission ( lxre ) of a `` golden sample '' ( gs ) of six long energetic grbs , when observed in the rest - frame of the source .
this remarkable feature , independent of the different isotropic energy ( @xmath0 ) of each grb , has been used to estimate the cosmological redshift of some long grbs .
this analysis is extended here to a new class of 161 long grbs , all with @xmath1 erg .
these grbs are indicated as binary - driven hypernovae ( bdhne ) in view of their progenitors : a tight binary system composed of a carbon - oxygen core ( co@xmath2 ) and a neutron star undergoing an induced gravitational collapse ( igc ) to a black hole triggered by the co@xmath2 explosion as a supernova ( sn ) .
we confirm the universal behavior of the lxre for the `` enlarged sample '' ( es ) of 161 bdhne observed up to the end of 2015 , assuming a double - cone emitting region .
we obtain a distribution of half - opening angles peaking at @xmath3 , with a mean value of @xmath4 , and a standard deviation of @xmath5 .
this , in turn , leads to the possible establishment of a new cosmological candle . within the igc model ,
such universal lxre behavior is only indirectly related to the grb and originates from the sn ejecta , of a standard constant mass , being shocked by the grb emission .
the fulfillment of the universal relation in the lxre and its independence of the prompt emission , further confirmed in this article , establishes a crucial test for any viable grb model . |
all graphs throughout are assumed to be finite , simple , and undirected . for terminology
not defined here we refer to @xcite or @xcite .
a graph is called _ @xmath0-connected _ ( @xmath12 ) if @xmath13 and @xmath14 is connected for all @xmath15 with @xmath16 .
let @xmath17 denote the _ connectivity _ of @xmath1 , that is , the largest @xmath0 such that @xmath1 is @xmath0-connected . a set @xmath15 is called a _
smallest separating set _ if @xmath18 and @xmath14 is disconnected . by @xmath19
we denote the set of all smallest separating sets of @xmath1 .
an edge @xmath3 of a @xmath0-connected graph @xmath1 is called _ @xmath0-contractible _ if the graph @xmath2 obtained from @xmath1 by _ contracting _
@xmath3 , that is , identifying its endvertices and simplifying the result , is @xmath0-connected . no edge in @xmath5
is @xmath0-contractible , whereas all edges in @xmath20 are if @xmath21 , and it is well - known and straightforward to check that , for a noncomplete @xmath0-connected graph @xmath1 , an edge @xmath3 is not @xmath0-contractible if and only if @xmath22 and @xmath23 for some @xmath24 .
there is a rich literature dealing with the distribution of @xmath0-contractible edges in @xmath0-connected graphs ( see the surveys @xcite ) , with a certain emphasis on the case @xmath25 . in @xcite ,
@xmath4-connected graphs that admit a spanning tree without any @xmath4-contractible edge have been introduced ; these were called _ foxes _ ( see figure [ fig : tightness ] ) .
for example , every wheel @xmath1 is a fox , which is certified by the spanning star @xmath26 that is centered at the hub of the wheel .
however , @xmath26 is as far from being a _ dfs _ ( depth - first search ) tree as it can be , and one could ask if the property of being a fox can be certified by some dfs tree at all .
the answer is no , as it has been shown in @xcite that every dfs tree of every @xmath4-connected graph nonisomorphic to @xmath10 does contain a @xmath4-contractible edge .
here we generalize the latter result as follows .
[ t1 ] every dfs tree of every @xmath4-connected graph nonisomorphic to @xmath10 , the prism @xmath27 , or the unique graph @xmath28 obtained from @xmath27 by adding a single edge contains at least two @xmath4-contractible edges .
theorem [ t1 ] is best possible in the sense that there is an infinite class of @xmath4-connected graphs admitting a dfs tree with only two @xmath4-contractible edges ( see figure [ fig : largedfstreewith2contractibleedges ] ) .
our proofs are based on methods introduced by mader in @xcite , generalizing the concept of critical connectivity .
this approach makes it possible to generalize some of the earlier results on foxes from @xmath4-connected graphs to certain classes of @xmath0-connected graphs . extending the definition above ,
let us define a _
@xmath0-fox _ to be a @xmath0-connected graph admitting a spanning tree without @xmath0-contractible edges . for @xmath29
, there are graphs @xmath1 without @xmath0-contractible edges at all , and every such @xmath1 is , trivially , a @xmath0-fox ; thus , the question is interesting only under additional constraints to @xmath1 which force @xmath0-contractible edges .
classic constraints are to forbid triangles or to bound the vertex degrees from below : in @xcite it has been proven that every triangle free @xmath0-connected graph contains a @xmath0-contractible edge , and in @xcite , it has been shown that every @xmath0-connected graph of minimum degree at least @xmath30 must contain a @xmath0-contractible edge ( unless @xmath1 is isomorphic to @xmath5 when @xmath31 ) . these results do have a common root in terms of generalized criticity @xcite , and so it is perhaps not surprising that the following new result , theorem [ t2 ] , follows from a statement on special separating sets ( theorem [ t5 ] in section [ sgeneral ] ) .
[ t2 ] let @xmath1 be a @xmath0-connected graph ( except for @xmath5 if @xmath6 ) that is triangle free or of minimum degree at least @xmath7 .
then every spanning tree of @xmath1 contains at least two @xmath0-contractible edges .
this implies that @xmath0-foxes must contain triangles as well as vertices of `` small '' degree . in order to show that the bound in theorem [ t2 ] is best possible
, we exhibit @xmath0-connected graphs of minimum degree @xmath9 ( and necessarily containing triangles ) that admit a spanning tree with no @xmath0-contractible edge . for odd @xmath32 ,
take the lexicographic product of any cycle and @xmath33 and add an additional vertex plus all edges connecting it to the others .
( so for @xmath25 we get back the wheels . )
the resulting graph is @xmath0-connected and of minimum degree @xmath9 , and the spanning star centered at the additional vertex has no @xmath0-contractible edge . the same construction works , more generally , if instead of a cycle we start with any _ critically @xmath34-connected graph _ , that is , a @xmath34-connected graph @xmath1 such that for every vertex @xmath35 the graph @xmath36 is not @xmath34-connected .
however , for dfs trees the situation changes once more : [ t3 ] for @xmath8 , every dfs tree of every @xmath0-connected graph of minimum degree at least @xmath9 contains at least two @xmath0-contractible edges .
observe that the statement of theorem [ t3 ] remains true for @xmath25 by theorem [ t1 ] unless the graph is one of the three exceptions listed there .
theorem [ t2 ] provides a particularly simple proof that every spanning tree of a _ cubic _ @xmath4-connected graph nonisomorphic to @xmath10 or the prism has at least two @xmath4-contractible edges ( see corollary [ c2 ] in section [ sgeneral ] ) ; however , taking more external knowledge into account we can improve _ two _ to the following sharp linear bound in terms of @xmath37 ( end of section [ sgeneral ] ) .
[ t4 ] every spanning tree of every cubic @xmath4-connected graph nonisomorphic to @xmath10 contains at least @xmath11 many @xmath4-contractible edges .
the bound is sharp , also when restricted to dfs trees .
we also show sharpness for theorem [ t4 ] .
obtain a graph @xmath38 from any cubic @xmath4-connected graph @xmath1 by replacing every vertex @xmath35 with a triangle @xmath39 such that , for every incident edge @xmath3 of @xmath35 , the end vertex @xmath35 of @xmath3 is replaced with a unique vertex of @xmath39 . clearly , @xmath38 is cubic and @xmath4-connected .
let @xmath40 be a spanning tree of @xmath1 , and let @xmath41 be formed by all edges of @xmath40 together with the edges of a spanning path of each @xmath39 .
then @xmath41 is a spanning tree of @xmath38 with exactly @xmath42 many @xmath4-contractible edges , as no edge in a triangle is @xmath4-contractible .
when restricted to dfs trees , assume in addition that @xmath1 is hamiltonian and let @xmath40 be a hamiltonian path of @xmath1 .
then the paths of each @xmath39 can be chosen such that @xmath41 is a hamiltonian path of @xmath38 and we see that there is no improvement for dfs trees in general .
let @xmath1 be a graph and @xmath19 be the set of its smallest separating sets . for @xmath43 , the union of the vertex sets of at least one but not of all components of @xmath14
is called a _
@xmath40-fragment_. obviously , if @xmath44 is a @xmath40-fragment then so is @xmath45 , where the index @xmath1 is always omitted as it will be clear from the context .
moreover , @xmath46 .
fragments have the following fundamental property . _
@xcite _ [ l1 ] let @xmath47 be a @xmath48-fragment and @xmath44 be a @xmath40-fragment of a graph @xmath1 such that @xmath49
. then @xmath50 , and if equality holds then @xmath51 is a @xmath52-fragment .
* let @xmath53 and observe that @xmath54 separates @xmath1 and is a subset of @xmath55 . therefore , @xmath56 .
since @xmath0 cancels on both sides , rearranging the terms yields the desired inequality , and equality implies @xmath57 .
@xmath58 we will not give explicit references to lemma [ l1 ] , but mark estimations or conclusions based on it by @xmath59 ; for example , we write @xmath60 if @xmath44 is a @xmath40-fragment and @xmath61 is a @xmath41-fragment such that @xmath62 to indicate that the inequality is a straightforward application of lemma [ l1 ] .
this convention also applies to the following slightly more complex but standard application of lemma [ l1 ] : if both @xmath51 and @xmath63 are nonempty , then , by lemma [ l1 ] , they are both fragments . in many cases , @xmath47 will be an inclusion minimal fragment with respect to some property , @xmath44 will be a @xmath40-fragment such that @xmath40 contains a vertex from @xmath47 , and @xmath64 will have the same property as @xmath47 ( but is no fragment by minimality ) : in such a scenario , we infer @xmath65 , @xmath66 , and @xmath67 from lemma [ l1 ] , and again refer to it by @xmath59 , for example , by writing @xmath68 , @xmath69 , or @xmath70 , respectively .
another fact that will be used throughout is that if @xmath44 is a @xmath40-fragment contained in some smallest separating set @xmath41 , then @xmath40 contains a vertex of every @xmath41-fragment ( as every vertex from @xmath71 must have a neighbor in every @xmath41-fragment , which can only be in @xmath40 ) ; since every component of @xmath72 is adjacent to all vertices of @xmath40 , it must contain a vertex from @xmath41 , so that , in particular , @xmath73 . now let us fix a subset @xmath74 of the power set @xmath75 .
we call a @xmath40-fragment @xmath44 a _ @xmath40-@xmath74-fragment _ if @xmath76 for some @xmath77 . in that case , again , @xmath72 is a @xmath40-@xmath74-fragment , too ; @xmath44 is called a _ @xmath40-@xmath74-end _
if there is no @xmath41-@xmath74-fragment properly contained in it , and @xmath44 is called a _ @xmath40-@xmath74-atom _ if there does not exist a @xmath41-@xmath74-fragment with fewer than @xmath78 vertices .
observe that if @xmath44 is a @xmath40-fragment then necessarily @xmath79 , so that @xmath40 can be reconstructed from @xmath44 ; therefore , one might omit @xmath40 in the notion , which defines the terms _ fragment _ , _ @xmath74-fragment _ , _
@xmath74-end _ , and
_ @xmath74-atom_. these definitions and the following theorem are from @xcite and have their roots back in a 1970 paper by watkins where it was proven that the degrees of a vertex transitive @xmath0-connected graph are at most @xmath7 @xcite . _
@xcite _ [ tmadergeneral ] let @xmath1 be a graph , @xmath80 , and @xmath81 be a @xmath82-@xmath74-atom of @xmath1 .
suppose that there exists an @xmath77 and a @xmath43 such that @xmath83 and @xmath84
. then @xmath85 and @xmath86 . a fragment of minimum size is usually called an _ atom _ of @xmath1 .
consequently , for @xmath87 , we obtain the following specialization of theorem [ tmadergeneral ] , which appeared already in @xcite . _
@xcite _ [ tmaderspecial ] let @xmath1 be a graph and @xmath81 be a @xmath82-atom of @xmath1 .
suppose that @xmath88 for some @xmath43
. then @xmath85 and @xmath89 .
we start our considerations with the following result .
[ t5 ] let @xmath26 be a spanning tree of a graph @xmath1 of connectivity @xmath0 , set @xmath90 , suppose that all @xmath74-fragments have cardinality at least @xmath91 , and let @xmath47 be an @xmath74-end
. then @xmath92 ( in particular , @xmath0 is odd ) or all edges @xmath3 from @xmath26 with @xmath93 are @xmath0-contractible .
* let @xmath94 and @xmath95 .
observe that the existence of @xmath47 implies @xmath96 .
since @xmath26 is a spanning tree , there exists an edge @xmath3 with @xmath93 .
if all such edges @xmath3 are @xmath0-contractible then we are done .
otherwise , one such edge @xmath3 is not @xmath0-contractible ; there exists a @xmath43 with @xmath23 , and we consider a @xmath40-fragment @xmath44
. now @xmath47 and @xmath44 are @xmath74-fragments , so that @xmath97 , and it suffices to prove that @xmath98 .
observe that @xmath99 , so @xmath100 .
if @xmath101 , we infer @xmath102 and @xmath103 , and , hence , @xmath104 , which is a contradiction .
suppose that @xmath49 .
then @xmath105 , which implies @xmath106 .
if @xmath107 , then @xmath108 , and otherwise @xmath109 , too . hence , @xmath110 , which is a contradiction
consequently , @xmath111 , and , by the same argument , @xmath105 . if @xmath112 , we infer @xmath113 , and otherwise @xmath114 , too .
symmetrically , we get @xmath115 and conclude @xmath116
. now @xmath112 implies @xmath117 , @xmath118 implies @xmath119 , and otherwise @xmath120 implies @xmath109 , too .
it follows @xmath121 , which proves the theorem .
@xmath58 [ c1 ]
let @xmath122 be a @xmath0-connected graph in which every fragment has cardinality at least @xmath123 .
then every spanning tree of @xmath1 admits at least two @xmath0-contractible edges
. * proof .
* let @xmath26 be a spanning tree of @xmath1 .
then @xmath124 , since @xmath125 .
hence , we may assume that at least one edge in @xmath26 is not @xmath0-contractible .
therefore , @xmath22 , and there exists an @xmath74-fragment @xmath126 , where we define @xmath90 as before .
in particular , @xmath96 .
consider an @xmath74-end @xmath127 . by theorem [ t5 ]
, @xmath26 contains a @xmath0-contractible edge @xmath3 that has precisely one end vertex in @xmath126 .
likewise , consider an @xmath74-end @xmath128 . by applying theorem [ t5 ]
once more , @xmath26 contains a @xmath0-contractible edge @xmath129 that has precisely one end vertex in @xmath130 . clearly , @xmath131 , which proves the statement .
@xmath58 for @xmath132 ,
the fragment size condition of corollary [ c1 ] is trivially true , and so every spanning tree of every @xmath0-connected graph nonisomorphic to @xmath5 , @xmath132 , admits at least two @xmath34-contractible edges . for general @xmath0 ,
corollary [ c1 ] implies theorem [ t2 ] as follows , and the examples beneath the latter in the introduction show also that the bound on the fragment size in corollary [ c1 ] can not be improved .
* proof of theorem [ t2 ] . *
as argued above , the statement holds for @xmath132 , so let @xmath32 .
then @xmath133 , since @xmath1 is triangle free or of minimum degree at least @xmath7 .
if the minimum degree condition is satisfied , every fragment has cardinality at least @xmath123 , and applying corollary [ c1 ] gives the claim .
if @xmath1 is triangle free , let @xmath26 be a spanning tree of @xmath1 and let @xmath90 .
as in the proof of corollary [ c1 ] , we may assume that at least one edge in @xmath26 is not @xmath0-contractible , which implies @xmath22 and the existence of a @xmath74-fragment @xmath126 .
since @xmath1 is triangle free , every @xmath74-fragment contains two adjacent vertices , and considering the neighborhood of the two vertices of degree at least @xmath0 each implies that every @xmath74-fragment has in fact cardinality at least @xmath0 . by applying theorem [ t5 ] twice , as in the previous proof , we find two @xmath0-contractible edges @xmath131 in @xmath26 with end vertices in @xmath126 and @xmath130 , respectively .
@xmath58 as promised in the introduction we derive the following result from theorem [ t2 ] .
[ c2 ] every spanning tree of every cubic @xmath4-connected graph nonisomorphic to @xmath10 or the prism @xmath27 contains at least two @xmath4-contractible edges .
* we use induction on the number of vertices .
the induction starts for @xmath10 , so suppose that @xmath1 is a cubic @xmath4-connected graph on at least six vertices , and @xmath26 is a spanning tree of @xmath1 .
we may assume that @xmath1 contains a triangle @xmath134 , for otherwise the statement follows from theorem [ t2 ] , and we may assume that @xmath1 is not the prism . the edge neighborhood of any such triangle forms a matching of three @xmath4-contractible edges , at least one of which belongs to @xmath26 ; in fact , we may assume that exactly one edge of the edge neighborhood belongs to @xmath26 , for otherwise the statement is proven .
so suppose that @xmath3 is the only edge from @xmath26 in the edge neighborhood of @xmath134 .
if there was another triangle @xmath135 then , consequently , @xmath3 is the only edge from @xmath26 in the edge neighborhood of @xmath135 , too , implying that @xmath136 , so that @xmath1 is the prism @xmath27 , which is a contradiction .
so we may assume that @xmath134 is the only triangle in @xmath1 .
the graph @xmath137 obtained from @xmath1 by identifying the three vertices of @xmath134 and simplifying is not @xmath10 , as @xmath1 is not the prism , and @xmath137 is not the prism , as @xmath134 is the only triangle in @xmath1 and the prism has two vertex - disjoint triangles .
clearly , @xmath137 is cubic and @xmath138 is a spanning tree of @xmath1 . since no smallest separating set of @xmath1 contains two vertices of @xmath134 , @xmath137 is also @xmath4-connected . hence , by induction , @xmath138 contains two @xmath4-contractible edges of @xmath137 , and the two edges corresponding to these in @xmath1 are @xmath4-contractible in @xmath1 as one checks readily .
@xmath58 by using a powerful result on @xmath4-contractible edges in @xmath4-connected graphs from the literature we can improve corollary [ c2 ] to theorem [ t4 ] ( where the lower bound to the number of @xmath4-contractible edges is sharp ) .
* proof of theorem [ t4 ] . * from lemma 3 and lemma 4 in @xcite
, we get theorem 3 in @xcite , which implies , together with theorem 12 from @xcite , that _ every vertex in a @xmath4-connected graph is either contained in a triangle or on at least two @xmath4-contractible edges_. we first show that this implies for any @xmath4-connected _ cubic _ graph @xmath1 that its subgraph @xmath139 on @xmath140 formed by the non-@xmath4-contractible edges of @xmath1 is a clique factor ( that is , all components of @xmath139 are isolated vertices , or single edges , or triangles or @xmath10 in case that @xmath1 is @xmath10 ) : suppose that @xmath1 is not @xmath10 .
if @xmath35 has degree at least @xmath34 in @xmath139 , then @xmath35 is on a triangle @xmath134 in @xmath1 by the result mentioned above and this triangle is also in @xmath139 , whereas the edges from its edge neighborhood are not ( so @xmath35 is in a triangle component of @xmath139 ) .
now every spanning tree @xmath26 contains at most @xmath34 edges from every triangle in @xmath139 , so that it contains at most @xmath141 edges from @xmath139 . therefore
, @xmath26 contains at least @xmath142 @xmath4-contractible edges ( unless @xmath1 is @xmath10).@xmath58
again , we observe that the spanning tree in the sharpness example of corollary [ c1 ] and theorem [ t2 ] is far from being a dfs tree
. the following theorem will provide more insight into the distribution of @xmath0-contractible edges in spanning trees of graphs where the fragment lower bound @xmath91 from theorem [ t5 ] is sharp ( as opposed to the bound @xmath123 that is sharp in corollary [ c1 ] ) , and leads to a proof of theorem [ t3 ] and , in the next section , of theorem [ t1 ] .
[ t6 ] let @xmath26 be a spanning tree of a graph @xmath1 of connectivity @xmath0 , set @xmath90 , suppose that all @xmath74-fragments have cardinality at least @xmath91 , set @xmath143 , @xmath144 for some @xmath74-end @xmath145 , and let @xmath47 be an @xmath146-end .
then @xmath92 , or @xmath26 contains a @xmath0-contractible edge @xmath3 with at least one endvertex in @xmath47 , or @xmath147 contains an @xmath146-fragment of cardinality @xmath91 such that all edges from @xmath26 having exactly one vertex in common with it are @xmath0-contractible .
* let @xmath94 .
let us call an edge from @xmath26 _ green _ if it is not @xmath0-contractible .
* claim 1 . *
suppose that @xmath3 is a green edge and @xmath81 is an @xmath74-end with @xmath144 .
then @xmath148 ( so that @xmath81 is an @xmath74-atom , and @xmath149 is an integer ) , and @xmath85 for every @xmath43 with @xmath23 ( and there exists such a @xmath40 ) .
we get @xmath148 immediately from theorem [ t5 ] .
since @xmath3 is green , there exists a @xmath43 such that @xmath23 , and for every such @xmath40 we know @xmath150 and @xmath88 , so that @xmath85 follows from theorem [ tmadergeneral ] , proving claim 1 .
now let us call a green edge @xmath3 _ red _ if @xmath144 for some @xmath74-end @xmath81 ( see figure [ fig : largedfstreewith2contractibleedges ] for an example of this coloring in the special case @xmath25 ) .
let @xmath95 .
by definition of @xmath146 and claim 1 , there exists a red edge @xmath151 and an @xmath152-end @xmath153 with @xmath154 and @xmath155 .
since @xmath47 is an @xmath152-fragment ( every @xmath146-fragment is an @xmath152-fragment by definition ) , it must contain an @xmath74-end @xmath81 .
there exists an edge @xmath3 from @xmath26 with @xmath144 .
if @xmath3 is @xmath0-contractible , we are done ; thus we may assume that @xmath3 is green , so that @xmath3 is even red .
by definition and claim 1 , there exists a @xmath156 such that @xmath157 .
we now consider a @xmath40-fragment @xmath44 , which is , in fact , an @xmath146-fragment , and analyze the possible ways @xmath158 meet @xmath159 ( for example , @xmath160 , @xmath161 and @xmath162 in figure [ fig : largedfstreewith2contractibleedges ] ) .
again we may rule out that @xmath101 , as this would imply @xmath163 and @xmath164 , and hence @xmath165 , which gives a contradiction .
now assume that exactly one of @xmath51 and @xmath166 is nonempty , say , by symmetry of @xmath44 and @xmath72 , @xmath49 ( see figure [ fig : structure ] ) .
it follows @xmath167 .
consequently , @xmath168 and , hence , @xmath169 . if @xmath112 , then @xmath117 , and if otherwise @xmath170 , then @xmath109 , too .
now @xmath171 , so we get equality summand - wise , that is , @xmath172 , @xmath173 , and @xmath174 .
this implies @xmath175 , so that @xmath176 ( in particular , @xmath72 is an @xmath146-atom ) and @xmath177 .
moreover , since @xmath155 and @xmath178 , we conclude @xmath179 , as @xmath180 $ ] is connected . in the remaining case @xmath181
, i.e. @xmath182 , we essentially obtain the same picture ( figure [ fig : structure ] ) : if @xmath183 , then @xmath113 , and if otherwise @xmath184 , then @xmath114 , too ; symmetrically , @xmath115 . then @xmath185 implies @xmath117 , @xmath186 implies @xmath187 , and the remaining case @xmath188 implies @xmath109 , too
. now if @xmath189 , the theorem is proved , so let @xmath190 .
then @xmath191 , so that equality holds summand - wise , implying @xmath174 , @xmath192 , @xmath193 , and @xmath173 . by symmetry of @xmath44 and @xmath72 , we may assume that @xmath194 and hence @xmath195 and @xmath196 .
this implies @xmath197 , as otherwise @xmath198 gives a contradiction .
regardless whether @xmath51 is empty or not , we conclude @xmath174 , @xmath177 with cardinality @xmath199 , @xmath200 with cardinality @xmath199 , and @xmath201 is an @xmath146-atom with cardinality @xmath149 and thus also an @xmath74-atom .
we proceed with the general argument .
if all edges from @xmath26 that connect @xmath72 to @xmath40 are @xmath0-contractible , we are done .
so there exists a green edge @xmath202 with @xmath203 and @xmath204 and a @xmath205 with @xmath206 , and by claim 1 ( applied to @xmath202 and @xmath72 ) we obtain @xmath207 .
we discuss the possible locations of @xmath208 and will show that all are impossible , which proves the theorem .
observe that @xmath81 must have at least @xmath209 neighbors in @xmath72 , for otherwise @xmath210 would be a nonempty set with less than @xmath0 neighbors in @xmath1 , which contradicts the fact that @xmath1 is @xmath0-connected .
it follows @xmath211 .
if @xmath208 would be the vertex in @xmath212 , then @xmath202 would be a red edge with its endvertices in @xmath213 , certifying that @xmath81 is an @xmath146-fragment properly contained in the @xmath146-end @xmath47 , which is absurd .
if @xmath208 would be some vertex in @xmath81 , then @xmath202 would be a red edge , and by claim 1 ( applied to @xmath202 ) , we get @xmath214 . as @xmath47 is an @xmath146-end , @xmath41 and @xmath48 _ cross _
( i.e. @xmath215 ) , and @xmath41 contains at least one vertex from @xmath216 ; on the other hand , @xmath41 contains @xmath217 vertices from @xmath218 , so that @xmath41 contains exactly one vertex from @xmath216 .
consider a @xmath41-fragment @xmath61 that contains @xmath219 ( this exists , as @xmath220 $ ] is connected and does not intersect @xmath41 ) .
then @xmath48 must intersect @xmath221 , which contradicts @xmath222 .
it follows that , necessarily , @xmath223 .
suppose to the contrary that @xmath41 separates @xmath224 , that is , there exists a @xmath41-fragment @xmath61 such that @xmath225 and @xmath226 .
then @xmath41 must contain a vertex @xmath227 from @xmath81 , and @xmath48 and @xmath41 cross , so that @xmath41 separates @xmath228 , too .
thus , @xmath41 contains a vertex @xmath229 from @xmath153 .
since @xmath41 can not separate the end vertices of @xmath230 , we know that @xmath231 or @xmath232 . without loss of generality we may suppose that @xmath232 .
if @xmath233 , then @xmath234 would be an @xmath235-fragment , as @xmath236 .
since @xmath151 is a green edge with exactly one end vertex from the @xmath74-atom @xmath153 and contained in @xmath237 we know from claim 1 that @xmath238 , contradiction .
if @xmath239 , then @xmath240 would be an @xmath241-fragment containing @xmath230 , as @xmath242 .
this implies again @xmath238 , contradiction .
therefore , @xmath243 .
it follows @xmath244 , contradiction .
hence we have to assume that @xmath41 does not separate @xmath224 .
consequently , there exists a @xmath41-fragment @xmath61 such that @xmath245 . if @xmath246 , then @xmath247 , and if otherwise @xmath248 , then @xmath249 , too . on the other hand , @xmath250 ,
we now prove a condition that guarantees two @xmath0-contractible edges in any spanning tree .
[ l2 ] let @xmath26 be a spanning tree of a noncomplete graph @xmath1 of connectivity @xmath0 , set @xmath90 , suppose that all @xmath74-fragments have cardinality at least @xmath91 , set @xmath143 , @xmath144 for some @xmath74-end @xmath145 .
then @xmath26 contains two @xmath0-contractible edges or there is an @xmath146-end of cardinality @xmath91 such that @xmath26 contains no @xmath0-contractible edge with an endvertex from it .
* call a fragment @xmath126 _ small _ if @xmath251 and _ big _ otherwise ; call @xmath126 _ good _ if @xmath26 contains a @xmath0-contractible edge that has at least one endvertex in @xmath126 and _ bad _ otherwise .
call @xmath126 _ very good _ if all edges from @xmath26 having exactly one endvertex in @xmath126 are @xmath0-contractible . in this language ,
theorem [ t6 ] tells us that an @xmath146-end @xmath47 is small , or good , or @xmath147 contains a small , very good @xmath146-fragment .
we may assume that @xmath26 contains at least one non-@xmath0-contractible edge , so that there exists an @xmath74-fragment , and , hence , an @xmath74-end , say , @xmath81 .
take an @xmath74-end @xmath252 .
there exist edges @xmath253 with @xmath144 and @xmath154 ; clearly , @xmath254 , and we are done if both @xmath255 are @xmath0-contractible . in what remains
we thus may assume that there _ exists _
an @xmath146-fragment .
suppose that there exists a very good @xmath146-fragment @xmath126 .
consider an @xmath146-end @xmath128 .
if @xmath47 is good , then there exists a @xmath0-contractible edge @xmath3 having a vertex in common with @xmath47 and another one having a vertex in common with @xmath126 , which proves the lemma . hence @xmath47 is bad .
if @xmath47 is small , then the lemma is proved again , so we may assume that @xmath47 is big . by theorem [ t6 ]
, @xmath147 contains a ( small ) very good @xmath146-fragment @xmath256 .
as @xmath126 and @xmath256 are disjoint and their union is not equal to @xmath140 , @xmath26 contains two distinct @xmath0-contractible edges that are incident with vertices from @xmath126 or @xmath256 , which proves the lemma .
therefore , we may assume that there are no very good @xmath146-fragments .
consequently , by theorem [ t6 ] , every big @xmath146-end is good .
moreover , we may assume that every small @xmath146-end is good , for otherwise we are done . hence , _ every _
@xmath146-end @xmath47 is good , and , for any @xmath146-end @xmath126 contained in @xmath216 , @xmath26 contains two distinct @xmath0-contractible edges that have an endvertex in @xmath47 and @xmath126 , respectively , which gives the lemma .
@xmath58 .
we now specialize to dfs trees .
a _ dfs tree _ of some graph @xmath1 is a spanning tree @xmath26 with a prescribed _ root vertex _
@xmath257 such that for every vertex @xmath35 , any two @xmath35-branches are nonadjacent in @xmath1 , where an _
@xmath35-branch _ is the vertex set of any component of @xmath258 not containing @xmath257 .
now we are prepared to prove the following theorem , which implies theorem [ t3 ] immediately , as , in a graph of connectivity @xmath0 and of minimum degree at least @xmath9 , every fragment has cardinality at least @xmath91 .
[ t7 ] let @xmath26 be a dfs tree of a noncomplete graph @xmath1 of connectivity @xmath8 , and set @xmath90 .
( i ) if all @xmath74-fragments have cardinality at least @xmath91 , then @xmath26 contains at least one @xmath0-contractible edge .
( ii ) if all fragments have cardinality at least @xmath91 , then @xmath26 contains at least two @xmath0-contractible edges .
* first , let us assume that all @xmath74-fragments have cardinality at least @xmath91 .
set @xmath143 , @xmath144 for some @xmath74-end @xmath145 and observe that @xmath259 . by lemma [ l2 ]
, we may assume that there exists a @xmath48-@xmath146-end @xmath47 with @xmath92 such that @xmath26 contains no @xmath0-contractible edge with an endvertex from @xmath47 .
there exists an edge @xmath260 and a @xmath82-@xmath74-end @xmath81 such that @xmath144 and @xmath261 . by claim 1 in the proof of theorem [ t6 ]
, we see that @xmath262 and @xmath263 .
let @xmath35 be the vertex in @xmath212 .
observe that @xmath47 is an @xmath74-end ( as it is even an @xmath74-atom ) and consider any edge @xmath264 with @xmath265 ( there exists at least one such edge ) .
let @xmath208 be the vertex in @xmath266 . as @xmath129 is not @xmath0-contractible
, there exists a @xmath43 with @xmath267 .
according to claim 1 in the proof of theorem [ t6 ] , @xmath182 .
all vertices from @xmath47 are neighbors of @xmath81 ( for otherwise @xmath268 would be a separating vertex set of @xmath1 with less than @xmath0 vertices ) , that is , @xmath269 .
let @xmath44 be a @xmath40-fragment .
then @xmath185 implies @xmath270 and @xmath184 implies @xmath271 , too .
as the same holds for @xmath72 instead of @xmath44 , we deduce @xmath272 and @xmath273 .
it follows that @xmath274 or @xmath275 , for otherwise @xmath40 contains a vertex from @xmath81 , as @xmath276 $ ] is connected ; this vertex can only be @xmath208 , so that applying claim 1 of the proof of theorem [ t6 ] on @xmath129 and @xmath81 ( or alternatively using theorem [ tmadergeneral ] with @xmath277 ) implies @xmath85 , which gives the contradiction @xmath278 .
it follows that @xmath279 or @xmath280 and @xmath281 , so that @xmath282 ( see figure [ fig : send4 ] ) . since
@xmath264 has been chosen arbitrarily from the edge neighborhood of @xmath47 , we see that @xmath35 is the only neighbor of @xmath47 in @xmath26 . in particular , @xmath81 is an @xmath146-end ( certified by any @xmath129 as above ) , since @xmath283 .
let us take over the notion of small , big , good , bad , and very good fragments from the previous lemma ; then @xmath81 and @xmath47 are small and @xmath47 is bad .
if @xmath81 was bad , too , then by symmetry of @xmath81 and @xmath47 we see that @xmath282 is the only neighbor of @xmath81 in @xmath26 .
since @xmath276 $ ] and @xmath284 $ ] are connected , each of them contains at most one @xmath35-branch of @xmath26 ; if each of them contains an @xmath35-branch , these @xmath35-branches span @xmath276 $ ] and @xmath284 $ ] and , as every vertex in @xmath81 has a neighbor in @xmath47 and vice versa , the two @xmath35-branches are adjacent , which is a contradiction . hence , one of @xmath276,g[b]$ ] does not contain an @xmath35-branch , and , thus , contains the root of @xmath26 , whereas the other one is spanned by a single @xmath35-branch . by symmetry ,
let us assume that @xmath47 contains the root , and consider any vertex @xmath285 .
again , @xmath227 and the vertices in @xmath81 are in different @xmath35-branches , but @xmath227 must have at least one neighbor in @xmath81 , contradiction . therefore , we may assume from now on that @xmath81 is good . hence , @xmath26 contains a @xmath0-contractible edge @xmath286 with @xmath287 and @xmath288 ( see also figures [ fig:6vertices1contractibleedgebw2 ] and [ fig:6vertices1contractibleedgebw ] ) .
this proves ( i ) of the statement .
now let us assume , in addition , that _ all _ fragments have cardinality at least @xmath91 , and ( reductio ad absurdum ) that @xmath286 is the unique @xmath0-contractible edge of @xmath26 .
observe that all small fragments of @xmath1 are atoms , so that , by theorem [ tmaderspecial ] , they are either subsets of or disjoint from any smallest separating set .
let us apply theorem [ t6 ] to an arbitrary @xmath146-end @xmath289 ( see figure [ fig : send5 ] ) .
if @xmath290 was good , then @xmath26 would contain a @xmath0-contractible edge that has an endvertex in @xmath290 , and is thus distinct from @xmath286 , contradiction .
hence @xmath290 is bad .
if @xmath290 was big , then @xmath291 would contain a small , very good @xmath146-fragment @xmath126 by theorem [ t6 ] . since @xmath286 is the unique @xmath0-contractible edge in @xmath26
, we see that @xmath292 , and , as @xmath126 is an atom and @xmath293 , we infer @xmath294 by theorem [ tmaderspecial ] . since @xmath295 is not @xmath0-contractible , @xmath296 , which implies that @xmath126 contains no vertex from @xmath47 either . as @xmath283
, it follows @xmath297 .
consequently , @xmath35 is the only neighbor of @xmath298 in @xmath26 , as @xmath298 is bad , @xmath126 is very good and @xmath35 is the only neighbor of @xmath47 in @xmath26 .
now it is not possible to locate the root vertex of our dfs tree : it can not be in @xmath299 , because then we find adjacent vertices from @xmath81 and @xmath47 in distinct @xmath35-branches , it can not be in @xmath300 , because then we find adjacent vertices from @xmath47 and @xmath301 in distinct @xmath35-branches , and it can not be in @xmath47 , because then we find adjacent vertices from @xmath126 and @xmath302 in distinct @xmath35-branches contradiction .
therefore , @xmath290 is a small , bad @xmath146-end , just as @xmath47 .
we may infer just as before for @xmath47 that @xmath303 contains a small , good @xmath304-@xmath146-end @xmath153 .
since @xmath305 we see that @xmath81 and @xmath153 are disjoint .
as @xmath153 is good and @xmath286 is the only @xmath0-contractible edge , it follows @xmath306 , and , as @xmath153 is an atom having a vertex in common with @xmath82 , we see that @xmath307 by theorem [ tmaderspecial ] .
since @xmath153 contains @xmath308 , we infer for the same reason that @xmath153 does not contain @xmath35 , and , as @xmath309 $ ] is connected and @xmath310 , @xmath153 does not contain any vertex from @xmath47 .
it follows @xmath311 .
moreover , we get , as for @xmath312 above , a unique vertex @xmath313 in @xmath314 and infer that all non-@xmath0-contractible edges from @xmath26 with exactly one vertex from @xmath290 are incident with @xmath313 .
we claim that @xmath286 is the only edge from @xmath26 that connects @xmath81 and @xmath153 .
assume to the contrary there was another one , say , @xmath315 , with @xmath316 and @xmath317 .
since @xmath318 is not @xmath0-contractible , there exists a @xmath205 with @xmath319 , and from theorem [ tmaderspecial ] we get @xmath320 .
since @xmath214 , @xmath41 contains at least one vertex from @xmath298 and thus consists of the @xmath321 vertices from @xmath322 and another vertex from @xmath298 ; but then @xmath323 induces a connected subgraph of @xmath1 , so @xmath41 does not separate @xmath82 , contradiction .
observe that ( when the position of the root in @xmath26 is neglected ) , the situation is symmetric in @xmath324 and @xmath325 .
we have seen that @xmath326 , @xmath327 , @xmath328 , and @xmath329 .
it follows that @xmath330 .
let us again analyze the position of the root vertex @xmath257 of our dfs tree .
then the @xmath331-path in @xmath26 must enter @xmath35 using an edge incident to @xmath81 or @xmath47 ( * ) , for otherwise @xmath260 implies that the vertex @xmath332 from @xmath333 and any vertex in @xmath334 are in distinct adjacent @xmath35-branches ( contradiction ) .
likewise , the @xmath335-path in @xmath26 must enter @xmath313 using an edge connecting to @xmath153 or @xmath290 ( * ) . setting @xmath336
we thus see that @xmath337 , for otherwise the second last vertex of every @xmath338-path in @xmath26 is either @xmath35 or @xmath313 and its @xmath331- or @xmath335-subpath violates ( * ) or ( * ) , respectively .
furthermore , @xmath339 , for otherwise @xmath340 violates ( * ) , @xmath341 violates ( * ) if the @xmath335-path in @xmath26 does not use the edge @xmath286 and @xmath341 violates ( * ) if it does , and we get symmetric violations for @xmath342 and @xmath343 , respectively .
now we can deduce that @xmath344 $ ] is connected , that is , a subtree of @xmath26 .
suppose , to the contrary , that @xmath345 contains a @xmath227 such that there is no @xmath346-path in @xmath345 .
the @xmath346-path in @xmath26 therefore has to use vertices from @xmath347 , and , therefore , both @xmath35 and @xmath313 .
if @xmath35 is used last ( that is , @xmath313 is on the @xmath331-path ) , then ( * ) is violated , and otherwise ( * ) is . by symmetry of @xmath312 and @xmath348
, we may assume that @xmath257 is not in @xmath47 .
since @xmath349 , there exists a vertex @xmath350 ; @xmath351 has a neighbor @xmath352 , and @xmath353 can not be in different @xmath354-branches for any vertex @xmath354 , so that either @xmath227 is on the @xmath355-path in @xmath26 or @xmath351 is on the @xmath346-path in @xmath26 .
the first option can not occur , as @xmath47 has only one neighbor in @xmath26 , and the second one implies that @xmath351 , like _ all _ vertices from the @xmath346-path , is in @xmath356 . since @xmath357 and @xmath358 , we deduce @xmath359 , in other words , @xmath360 . by theorem [ tmaderspecial ] , @xmath361 , which determines @xmath362 , @xmath363 , @xmath364 , and @xmath365 .
the set @xmath366 has all its neighbors in @xmath367 , which implies that @xmath368 is empty .
therefore , @xmath369 , so that @xmath370 . hence @xmath371 is a fragment of cardinality @xmath372 , contradiction . @xmath58
for proving theorem [ t1 ] , observe that the fragment size conditions from theorem [ t7 ] are vacuously true in the case that @xmath25 .
however , the conclusion in ( ii ) does not hold for that case , as shown for its direct implication theorem [ t3 ] in the introduction , so we may expect some differences in the argumentation for @xmath25 .
nonetheless , a substantial part of the proof of theorem [ t6 ] can be taken over .
* proof of theorem [ t1 ] .
* let @xmath26 be a dfs tree of a @xmath4-connected graph @xmath1 nonisomorphic to @xmath10 .
we will prove that @xmath26 contains at least one @xmath4-contractible edge , and that @xmath26 contains at least two @xmath4-contractible edges unless @xmath1 is a prism or a prism plus a single edge ( and @xmath26 has a special shape ) .
set @xmath90 and @xmath143 , @xmath144 for some @xmath74-end @xmath145 .
let us adopt once more the terminology of small , big , good , bad , and very good fragments introduced in the proof of lemma [ l2 ] . by lemma [ l2 ] ( which is , in contrast to theorem [ t7 ] , true for @xmath25 , too )
, we may assume that there exists a small , bad @xmath48-@xmath146-end @xmath373 .
there exists an edge @xmath260 and a @xmath82-end @xmath81 such that @xmath261 and @xmath144 , and as in claim 1 in the proof of theorem [ t6 ] we see that @xmath374 , say , @xmath375 .
let @xmath35 be the vertex in @xmath376 and let @xmath351 be the vertex in @xmath377 . as @xmath378 is @xmath4-contractible ( because @xmath379 induces a triangle in @xmath1 ) , we know that @xmath380 , as otherwise @xmath47 would not be bad . hence , as @xmath381 , exactly one of @xmath382 is in @xmath383 .
observe that @xmath149 is not adjacent to @xmath351 , because @xmath149 has degree @xmath4 and a neighbor in @xmath216 .
suppose that @xmath384 .
then there exists a @xmath43 containing @xmath385 , and a vertex @xmath332 from @xmath216 .
consider any @xmath40-fragment @xmath44 .
at least one of @xmath386 must be empty , for otherwise one of these sets is an @xmath387-fragment and the other one is an @xmath388-fragment ; this is not possible , because @xmath149 has only one neighbor in @xmath216 .
so @xmath389 or @xmath390 is a @xmath40-fragment , but the latter is not because @xmath351 is not adjacent to @xmath391 .
consequently , @xmath389 is a @xmath40-fragment , and , in fact , an @xmath74-atom .
therefore , after possibly having exchanged the names of @xmath392 ( and resetting @xmath81 accordingly ) , we may assume without loss of generality that @xmath393 ; @xmath82 consists of @xmath394 , and a vertex @xmath332 from @xmath216 , and we see immediately that @xmath375 is a small @xmath146-end , too .
the edge @xmath395 is 3-contractible , because @xmath396 induces a complete graph in @xmath1 .
now if the small @xmath146-end @xmath81 was bad , @xmath397 would be the only edge incident with @xmath149 in @xmath26 , since @xmath395 is @xmath4-contractible .
then one of @xmath149 and @xmath398 must be the root vertex @xmath257 of the dfs tree @xmath26 , because otherwise @xmath399 would belong to different @xmath35-branches but are adjacent .
if @xmath400 , then @xmath398 and @xmath351 belong to different @xmath35-branches but are adjacent ; if otherwise @xmath401 , then @xmath149 and @xmath332 belong to different @xmath35-branches but are adjacent .
thus , @xmath81 is good .
since @xmath402 and @xmath397 is not @xmath4-contractible , it follows that @xmath26 contains the @xmath4-contractible edge @xmath395 .
this gives the first claim ; for the second , let us assume that @xmath395 is the only @xmath4-contractible edge from @xmath26 .
we have to prove that @xmath1 is either the prism or the prism plus a single new edge ; to this end , we proceed as in the proof of theorem [ t7 ] , and consider an arbitrary @xmath403-@xmath146-end @xmath289 .
if it was good , we would find a @xmath4-contractible edge having an endvertex in common with @xmath290 , and , thus , distinct from @xmath395 , which gives a contradiction .
thus , @xmath290 is bad .
if @xmath290 was big , it would contain a very good end in its neighborhood ; just as in the proof of theorem [ t7 ] , we can not locate the root of @xmath26 properly , which gives a contradiction . therefore , @xmath290 is a bad , small @xmath146-end , and we may apply to @xmath290 the same line of arguments that we applied before to @xmath47 : @xmath403 consists of three vertices @xmath404 , where @xmath405 is a good , small @xmath304-@xmath146-end , and @xmath304 consists of @xmath406 and a vertex @xmath407 from @xmath408 , where , moreover , @xmath409 and @xmath410 are from @xmath383 and not @xmath4-contractible , and @xmath411 is from @xmath383 and @xmath4-contractible . since @xmath395 is the only @xmath4-contractible edge , we see that @xmath412 , which implies @xmath413 and @xmath414 .
again the situation is symmetric in @xmath415 and @xmath416 , and we may proceed almost literally as in the proof of theorem [ t7 ] by showing first that @xmath339 , @xmath417 , and @xmath418 $ ] is a subtree of @xmath26 , and thus , more precisely , a path @xmath419 . by symmetry , we may assume that @xmath420 , and , as @xmath398 can not be on the @xmath355-path in @xmath26 , that @xmath351 must be on the @xmath421-path in @xmath26 and hence in @xmath422 .
different from the more general argument , we now have two options for @xmath351 : @xmath351 is either @xmath423 , or it is @xmath313 .
we consider the corresponding cases separately : if there are no further vertices but @xmath429 and they are all distinct , then @xmath430 . as certified by the @xmath74-end @xmath431 and the edge @xmath432
, @xmath433 is an @xmath146-end contained in @xmath298 .
as we have seen above ( for the arbitrarily chosen @xmath290 ) , the neighborhood of such a fragment necessarily contains @xmath434 , contradiction .
if there are further vertices other than those listed in the previous paragraph , they are all from @xmath426 . then @xmath427 has all its neighbors in @xmath435 .
therefore , @xmath436 , and @xmath427 is , indeed , an @xmath40-fragment , and @xmath437 . since @xmath438 has only one neighbor in @xmath439 , @xmath440 is an @xmath441-fragment ( with @xmath442 ) .
the @xmath74-end @xmath431 together with the edge @xmath432 certifies that @xmath44 is an @xmath146-fragment , and , thus , contains an @xmath146-end @xmath443 , which is in @xmath298 .
however , the neighborhood of @xmath443 does not contain @xmath434 , as it should , by what we have proven about the arbitrarily chosen @xmath290 above , which gives a contradiction .
therefore , there are no further vertices but @xmath429 , and the vertices listed are not all distinct .
this implies @xmath444 .
the neighborhoods of @xmath445 are determined , so that @xmath446 is determined up to a possible edge connecting @xmath35 and @xmath313 .
if @xmath35 and @xmath313 are not adjacent , we get the prism , and otherwise we get the prism plus a single edge .
99 and u. s. r. murty , `` graph theory '' , graduate texts in mathematics 244 ( 2007 ) , springer london .
, `` graph theory '' , 4th edition , graduate texts in mathematics 173 ( 2010 ) , springer heidelberg .
, `` contractible edges in @xmath447-connected graphs with minimum degree greater than or equal to @xmath448 $ ] '' , graphs and combinatorics 7 ( 1990 ) , 1521 . , k. mehlhorn , and j. m. schmidt , `` every dfs tree of a @xmath4-connected graph contains a contractible edge '' , j. graph theory 72 ( 2013 ) , 112121 . , ``
minimal connectivity '' , in : `` topics in structural graph theory '' , edited by l. beineke and r. wilson , cambridge university press ( 2012 ) .
, `` a survey on contractible edges in graphs of a given vertex connectivity '' , graphs and combinatorics 18 ( 2002 ) , 130 . , `` contractible subgraphs in 3-connected graphs '' , j. combin .
theory ser .
b 80 ( 2000 ) , 3248 . , `` generalizations of critical connectivity of graphs '' , discrete math . 72 ( 1988 ) , 267283 . , `` eine eigenschaft der atome endlicher graphen '' , arch . math .
( basel ) 22 ( 1971 ) , 333336 .
, `` nonseparating cycles in @xmath0-connected graphs '' , j. graph theory 5 ( 1981 ) , 351354 . , `` connectivity of transitive graphs '' , j. combin .
theory 8 ( 1970 ) , 2329 . | an edge in a @xmath0-connected graph @xmath1 is called _ @xmath0-contractible _ if the graph @xmath2 obtained from @xmath1 by contracting @xmath3 is @xmath0-connected .
generalizing earlier results on @xmath4-contractible edges in spanning trees of @xmath4-connected graphs , we prove that ( except for the graphs @xmath5 if @xmath6 ) ( a ) every spanning tree of a @xmath0-connected triangle free graph has two @xmath0-contractible edges , ( b ) every spanning tree of a @xmath0-connected graph of minimum degree at least @xmath7 has two @xmath0-contractible edges , ( c ) for @xmath8 , every dfs tree of a @xmath0-connected graph of minimum degree at least @xmath9 has two @xmath0-contractible edges , ( d ) every spanning tree of a cubic @xmath4-connected graph nonisomorphic to @xmath10 has at least @xmath11 many @xmath4-contractible edges , and ( e ) every dfs tree of a @xmath4-connected graph nonisomorphic to @xmath10 , the prism , or the prism plus a single edge has two 3-contractible edges .
we also discuss in which sense these theorems are best possible . *
ams classification : * 05c40 , 05c05 .
* keywords : * contractible edge , spanning tree , dfs tree , fox . |
supernova remnants ( snrs ) are canonically considered the main sources of cosmic - rays in the galaxy .
the detection of non - thermal x - ray emission from snrs clearly indicates that electrons are actually accelerated around the snr shocks @xcite , and the observations can constrain the electron spectra . on the other hand , observational confirmation of accelerated protons is not as easy as that of electrons .
one way to study the acceleration and spectrum of protons is to study gamma - ray emission through @xmath2-interactions and the decay of neutral pions ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
in particular , molecular clouds are efficient targets of cosmic - ray protons because of their high density .
thus , clouds illuminated by the protons accelerated in a nearby snr could be bright gamma - ray sources ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
theoretical studies have suggested that old snrs could be appropriate objects to investigate gamma - ray emission through @xmath2-interactions , because the radiation from the accelerated electrons ( primary electrons ) disappears as the snr evolves , owing to their short cooling time @xcite .
in other words , we could ignore the gamma - rays from primary electrons via inverse - compton ( ic ) scattering of ambient soft photon fields and/or non - thermal bremsstrahlung from the interaction of electrons with dense ambient matter . in this letter
, we consider the evolution of an snr surrounded by molecular clouds .
we calculate the spectrum of cosmic - rays accelerated in the snr and the photon spectrum of the molecular clouds illuminated by the cosmic - rays .
we assume that a supernova explodes in a low - density cavity , because the progenitor star expels ambient dense gas via strong uv - radiation and stellar winds @xcite .
the cavity may also have been created through previous supernova explosions .
what differentiates this study is that we consider whether high - energy cosmic - rays illuminating molecular clouds were accelerated even after the snr became old or they were accelerated only when the snr was young .
we also discuss the influence of particle diffusion on the cosmic - ray spectrum .
we construct specific models for the open cluster westerlund 2 and the snr w 28 , and compere the results with latest observations .
westerlund 2 is one of the young open clusters from which tev gamma - ray emission has been detected with h.e.s.s .
it is surrounded by molecular clouds @xcite .
the gamma - ray emission is extended ( @xmath3 ) and covers the molecular clouds @xcite .
noticeable objects such as pulsars that can be the source of the gamma - ray emission have not been observed in this region .
@xcite proposed that the gamma - ray emission comes from an old snr , although there is no clear signature of snrs in the cluster .
w 28 is a mixed - morphology snr interacting with a molecular cloud @xcite .
it is an old snr and tev gamma - rays have been detected from molecular clouds around the snr @xcite .
as will be shown in the next section , the proton spectrum around westerlund 2 and w 28 can be fitted with a power - law with an index @xmath4 ( the definition is shown in equation [ [ eq : npe ] ] ) . moreover
, the shock waves seem to have finished accelerating particles for those objects , while the surrounding regions are bright in the gamma - ray band . here
, we briefly discuss what they mean before we explain assumptions in our specific models for westerlund 2 and w 28 .
if the duration of particle acceleration is shorter than that of the diffusion , and the particle source is spatially localized well , we can use the analytical solution in @xcite .
this corresponds to the situation where particles are accelerated mainly when the snr is young and compact , and the molecular cloud illuminated by accelerated cosmic - rays is distant from the snr shock .
if the shape of the source spectrum is a power - law with an index @xmath5 or @xmath6 , the energy spectrum at the position of the cloud ( @xmath7 ) is represented by @xmath8 if radiative cooling during the diffusion can be ignored .
the diffusion length is @xmath9 , where @xmath10 is the diffusion coefficient .
following @xcite , we assume that @xmath11 where @xmath12 is the magnetic field . at a point distant from an snr
, we expect that @xmath13 .
thus , for a given magnetic field , the energy spectrum is represented by @xmath14 if @xmath15 at the position of the molecular cloud .
this means that even if particles are accelerated efficiently ( @xmath16 ) , the energy spectrum must be soft ( @xmath17 ) . in other words , if the index of the spectrum is observed to be @xmath18 at a molecular cloud , it is likely that the particles are accelerated near the cloud after the snr becomes large . for westerlund 2 and w 28 , since the spectral indices are @xmath4 for the high - energy protons
( @xmath19 tev ) that are illuminating the molecular clouds around these objects , we expect that the cosmic - rays were accelerated near the molecular clouds even after the snrs became old and large .
this may be in contrast with the assumption often adopted in theoretical studies .
we assume that the snr shock had traveled in a low - density cavity . during the propagation in the cavity ,
the shock wave is in the adiabatic sedov phase , because the low - density prevents the radiative cooling .
thus , even if particles can be accelerated only during the sedov phase , they are being accelerated until the shock reaches and collides with the surrounding high - density region , which is an effective target of the cosmic - rays .
the particles illuminate the high - density region with the energy spectrum at the acceleration site or the shock .
thus , the spectral indices of @xmath18 are possible . for westerlund 2 and w 28 , efficient particle acceleration around shock waves seems to have finished . for the former ,
there is no signature of a shock .
thus , the shock may have already collided with the surrounding dense region and dissipated . for
the latter , the shock is traveling in a relatively high density region ( @xmath20@xmath21 ) and seems to be in a radiative phase @xcite .
thus , cosmic - rays that are illuminating molecular clouds might be accelerated in the past . for @xmath22 and @xmath23 in equation ( [ eq : de ] ) , @xcite estimated the diffusion time of cosmic - rays in a typical molecular cloud and showed that it is only @xmath24 yr for protons with an energy of @xmath20 tev .
however , the probability that the particle acceleration has finished within @xmath24 yr must be very small .
therefore , in the next section we simply assume that @xmath25 around the snrs and later discuss it quantitatively
. the small @xmath26 may be because of the generation of plasma waves by cosmic - rays @xcite .
note that if particles are accelerated with a spectral index of @xmath16 , and if they diffuse in the interstellar gas and illuminate molecular clouds before the index increases to @xmath27 , the observed spectral index could be @xmath4 as the ones observed in westerlund 2 and w 28 .
however , the value of @xmath26 and/or the distance between the snr and the molecular clouds must be fine - tuned in this case . in particular , for w 28 , the molecular clouds are adjacent to the snr @xcite and there seems to be no room to adjust the distance .
thus , we do not consider this possibility .
we assume that the distance to westerlund 2 is @xmath28 kpc @xcite .
we also assume that a supernova exploded at the center of a cavity , which is filled with warm gas , and is surrounded by molecular clouds ( fig .
[ fig : conf]a ) . after the explosion
, a shock expands in the warm gas and then hits the surrounding molecular gas ( @xmath29 pc from the cavity center ) .
the size of the cavity is based on the arc structure found by @xcite .
when the shock expands in the cavity , the shock velocity @xmath30 is written as a function of the snr age ( @xmath31 ) as @xmath32 where @xmath33 cm s@xmath34 is the initial velocity of the ejecta @xcite . at @xmath35
yr , the snr is in the free expansion phase , at @xmath36 yr , it is in the radiative phase , and at @xmath37 , it is in the sedov phase . here
@xmath38 and @xmath39 , where @xmath40 is the hydrogen density in the cavity and @xmath41 is the mean atomic weight of the gas assuming one helium atom for every 10 hydrogen atoms .
the initial energy of the ejecta is @xmath42 erg .
the radius of the shock ( @xmath43 ) can be obtained by integrating equation ( [ eq : vs ] ) .
the energy spectrum of the accelerated protons is given by @xmath44 the index is given by @xmath45 , where @xmath46 is the compression ratio of the shock @xcite , which is given by the rankine - hugoniot relation for @xmath37 .
we do not consider non - linear effects , and thus @xmath47 is equal to or smaller than 4 .
we assume that the maximum energy is determined by the age of the snr : @xmath48 where @xmath49 is the factor determined by the shock angle and the gyro - factor , and @xmath50 is the downstream magnetic field , which is given by @xmath51 , where @xmath52 is the magnetic field in the cavity @xcite .
the minimum energy of the protons is given by the rest - mass energy . in our fiducial model , we assume that @xmath53 , which is the lower end of the typical value in such regions ( @xmath54@xmath55 ; @xcite ) . there is no information about the temperature and magnetic field in the cavity before the explosion .
thus , we assume that @xmath56 k and @xmath57 so that the photon spectrum is consistent with observations ; @xmath58 and @xmath52 determine @xmath59 and @xmath60 , respectively .
it is to be noted that the current temperature of the warm gas is @xmath61@xmath62 k @xcite . for @xmath63 ,
the shock radius is @xmath64 pc at @xmath65 yr .
we define this time as @xmath66 and the shock collides with the molecular clouds around this time . since @xmath67 yr , the snr is at the end of the sedov phase at @xmath68 . assuming that particle acceleration stops at @xmath69 , the protons accelerated at @xmath70 ( @xmath71 )
illuminate the molecular clouds .
for @xmath72 , the compression ratio @xmath47 and the proton cut - off energy @xmath73 decrease from 4 to 3.6 and from 151 to 47 tev , respectively .
therefore , the protons that illuminate the molecular clouds have a spectrum represented by equation ( [ eq : npe ] ) with @xmath74 and @xmath75 tev .
the mass of the molecular clouds around westerlund 2 is @xmath76 @xcite , although the clouds are very irregular .
referring to the observations , we assume that the proton density of the clouds is @xmath77 . in the clouds
, we expect that the shock follows the `` snowplow '' evolution ( @xmath78 ) .
we assume that the shock in the clouds can locally be expressed as a shell centered on @xmath79 . in this case , from the momentum conservation , the radius of the shell , @xmath80 , has a relation of @xmath81\dot{r}_{\rm sh}(t ) = \frac{4\pi}{3}n_{\rm cav } r_s(t_{\rm coll})^3 v_s(t_{\rm coll } ) \:,\ ] ] with @xmath82 at @xmath83 , which can be solved numerically for @xmath78 .
after the collision , it takes @xmath84 yr until the shell velocity decreases to the internal velocity of the clouds ( @xmath85 ; @xcite ) .
since the shock has not been observed apparently , its velocity may have decreased to that level .
some of the accelerated protons plunge into the molecular clouds and create electrons , positrons , and gamma - ray photons through @xmath2-interactions there .
the electrons and positrons ( both are called secondary electrons ) also emit photons through synchrotron , ic , and bremsstrahlung radiation .
figure [ fig : wd2 ] shows the broad - band spectrum of the molecular clouds at @xmath86 .
we assume that the spectrum of the cosmic - ray protons is represented by equation ( [ eq : npe ] ) with @xmath87 and @xmath88 tev .
the total energy of the protons illuminating the molecular clouds is @xmath89 erg .
note that the total number of the target protons and the total energy of the illuminating protons are degenerated for the photon spectrum .
the magnetic field in the cloud is @xmath90 .
the spectrum is calculated based on the radiation models in @xcite .
for the @xmath2-interactions , we used the code provided by @xcite . in figure
[ fig : wd2 ] , we included the contribution of background cosmic - rays ( equation [ 10 ] in @xcite ) , but it is negligibly small .
we assumed that the injection of secondary electrons is balanced with the cooling .
the model results are compared with observations .
the radio flux is considered as an upper limit , because it could only partially include the non - thermal radio flux produced by the energetic electrons @xcite .
for the two - band fermi observations , we assume that the spectral index is between @xmath91 and @xmath92 . the data were particularly important to constrain the value of @xmath59 .
the diffusion time of cosmic - rays in a cloud is written as @xmath93 where @xmath94 is the size of a molecular cloud . from the condition of @xmath95
, the reduction factor must be @xmath96 .
if we ignore the diffusion of the protons in the cloud , the emission originated from the protons will not change for a long time because of the long cooling time of the protons ( @xmath97 yr ; @xcite ) . in figure
[ fig : wd2 ] , the emission from primary electrons is also presented .
their energy spectrum is given by @xmath98 the maximum energy is determined by synchrotron cooling : @xmath99 @xcite
. the minimum energy of the electrons is given by the rest - mass energy .
we assume that the electron - proton ratio is @xmath100 . if the bremsstrahlung radiation from primary electrons were more than ten times brighter than that in figure [ fig : wd2 ]
, it might overwhelm the gamma - ray emission from the pion decay .
however , the required electron - proton ratio ( @xmath101 ) would be unrealistically large .
moreover , because of the short cooling time , the radiation of the primary electrons should decrease after particle acceleration stops .
thus , for @xmath78 ( @xmath71 ) , observed flux from the primary electrons should be lower than that in figure [ fig : wd2 ] .
we apply the same model to w 28 .
a supernova explodes in a low - density cavity ( fig .
[ fig : conf]b ) .
the distance to w 28 has been estimated to be @xmath1023 kpc @xcite ; we assume @xmath103 kpc in this letter .
we assume that @xmath104 , @xmath105 , and @xmath106 k. for @xmath107 and @xmath108 @xcite , the shock radius is @xmath109 pc at @xmath110 yr . during the sedov phase , the index @xmath59 increases from 2.0 to 2.3 , and the maximum proton energy @xmath111 decreases from 8.8 to 2.7 tev . at @xmath69 , we assume that the shock collides with a surrounding relatively dense region of @xmath112 ( fig .
[ fig : w28]b ) and we define this time as @xmath83 . the density @xmath113 is based on observations @xcite .
the detection of 1720 mhz maser emission from w 28 actually indicates the presence of the shock in the dense gas @xcite . after the collision
, the evolution of the shock follows equation ( [ eq : mom ] ) in which @xmath114 is replaced by @xmath113 .
the radius of the shell increases from @xmath115 to 13 pc , and the velocity decreases from @xmath116 to @xmath117 in @xmath118 yr
. the radius of @xmath119 pc and the velocity of @xmath117 are comparable to the observed ones @xcite .
note that even if particles can be accelerated at radiative shocks , it can not be applied to w 28 at present , because the small velocity ( @xmath117 ) prevents ionization of the preshock gas @xcite . since the gamma - ray emission comes from clumps around the snr @xcite
, we expect that there are dense clumps , which are the targets of cosmic - ray protons ( fig .
[ fig : conf]b ) . referring to the observations
, we assume that the mass and density of a clump is @xmath120 and @xmath121 , respectively @xcite . figure [ fig : w28 ] shows the broad - band spectrum of the molecular cloud at @xmath86 .
the spectrum of the cosmic - ray protons is represented by equation ( [ eq : npe ] ) with @xmath87 and @xmath122 tev .
the total energy of the protons that contribute to the illumination of the molecular clouds is @xmath123 erg .
the magnetic field in the cloud is @xmath124 .
the radio flux is considered as an upper limit , because it includes radio emission outside of the region where tev gamma - ray has been detected @xcite . since the xmm - newton upper - limit is for part of the tev gamma - ray region , the actual upper - limit may be larger ( nakamura et al . 2009 , in preparation ) . as is the case for westerlund 2 , while the photon flux from primary electrons decreases as the time advances , that originated from protons does not .
the condition @xmath125 gives @xmath126 .
recently , fermi found gev gamma - rays in the molecular clouds where tev gamma - ray has been detected @xcite , which indicates that the energy dependence of @xmath10 is not affecting the proton spectrum and is consistent with our assumption that the protons were accelerated near the molecular clouds .
in this letter , we constructed a model of gamma - ray emission from molecular clouds surrounding an snr .
we considered old snrs because gamma - rays mainly come through @xmath2-interactions and pion decay , and those from primary electrons can be ignored .
in fact , the synchrotron cooling time of electrons with @xmath127 tev is only @xmath128 yr for @xmath129 @xcite , and thus they cool quickly after their acceleration finishes .
we assumed that the snr shock travels in a low - density cavity , and we indicated that the diffusion coefficient of cosmic - ray particles around the snr needs to be much smaller than that away from it ( less than @xmath0% ) , the predicted broad - band spectra for westerlund 2 and w 28 are consistent with observations ( figs .
[ fig : wd2 ] and [ fig : w28 ] ) .
the proton spectra can be fitted with a power law with indices @xmath1 , which is the value at cosmic - ray sources expected in cosmic - ray propagation models for the galaxy @xcite .
the indices suggest that particles were accelerated near the molecular clouds even after the snr became old and large for these objects
. the soft spectra may be because the particles were accelerated at the end of the sedov phase . while the low mach number of the shock makes the spectra soft in our model , neutral hydrogens in the cavity may also do that @xcite .
on the other hand , for another snr ic 443 , the broad - band photon spectrum indicates that the index is 2 @xcite .
for this object , protons accelerated at an earlier stage of the sedov phase ( when the mach number of the shock is large ) may be illuminating molecular clouds .
moreover , if molecular clouds are distributed with various sizes and distances from the snr , the gamma - ray spectrum may be the superposition of the spectrum of each cloud and may reflect the energy - dependence of particle diffusion ( equation [ [ eq : de ] ] ; see @xcite ) . | we study cosmic - ray acceleration in a supernova remnant ( snr ) and the escape from it .
we model nonthermal particle and photon spectra for the hidden snr in the open cluster westerlund 2 , and the old - age mixed - morphology snr w 28 .
we assume that the snr shock propagates in a low - density cavity , which is created and heated through the activities of the progenitor stars and/or previous supernova explosions .
we indicate that the diffusion coefficient for cosmic - rays around the snrs is less than @xmath0% of that away from them .
we compare our predictions with the gamma - ray spectra of molecular clouds illuminated by the cosmic - rays ( fermi and h.e.s.s . ) .
we found that the spectral indices of the particles are @xmath1 .
this may be because the particles were accelerated at the end of the sedov phase , and because energy dependent escape and propagation of particles did not much affect the spectrum . |
horn and weinstein@xcite and horn et al@xcite proposed the @xmath0expansion for the calculation of the ground
state energy of quantum mechanical models .
it is the taylor expansion about @xmath1 of a monotonically decreasing function @xmath2 that leads to the ground
state energy when @xmath3 provided that the chosen reference function exhibits a nonzero overlap with the ground state .
the coefficients of such maclaurin series are known as connected moments or cummulants .
since the extrapolation of the @xmath0expansion towards @xmath3 by means of pad approximants@xcite did not appear to produce encouraging results , cioslowski@xcite proposed an exponential series and stubbins@xcite compared it with other extrapolation approaches .
on matching the @xmath0expansion and the exponential series at origin one has to solve a system of nonlinear equations . in order to bypass this problem
cioslowski@xcite developed a systematic algorithm for obtaining just the parameter related to the energy by removing all the other variables in the nonlinear equations .
the resulting approach is known as connected moments expansion ( cmx ) . from the properties of the pad approximants
knowles@xcite derived a compact an ellegant explicit expression for the approximants to the energy in terms of the connected moments .
since the cmx approximants may exhibit singularities@xcite other authors proposed improved approaches like the alternate moments expansion ( amx)@xcite and the generalized moments expansion(gmx)@xcite .
although these methods may avoid the singularities in the cmx approximants they do not seem to improve the convergence properties of the sequences of approximants and commonly their results are not more accurate than the cmx ones . a recent analysis of the convergence properties of the cmx required the calculation of all the variables in the cmx ansatz@xcite .
for that reason , amore and fernndez@xcite proposed a systematic method for the full solution of the cmx nonlinear equations motivated by an earlier discussion@xcite of the connected moments polynomial approach@xcite .
nonlinear equations like the cmx ones are an old problem in applied mathematics and were first solved by prony@xcite and later weiss and mcdonough@xcite proposed an alternative approach based on pad approximants .
in fact , prony s method is well known in numerical analysis@xcite and has been widely applied to a variety of problems in chemistry and engineering@xcite .
further inspection of the procedure developed by amore and fernndez@xcite has revealed that it is closely related to prony s method , although the main result derived by the former authors does not appear in the discussions of the latter method@xcite .
the main purpose of this paper is to show the connection between prony s method and the procedure developed by amore and fernndez . in sec .
[ sec : prony ] we show how to solve a particular set of nonlinear equations by means of prony s method and derive the main result of amore and fernndez@xcite . in sec .
[ sec : generating ] we discuss the application of that method to the extrapolation of the generating functions for the moments and connected moments . in sec .
[ sec : examples ] we test the general analytical results on simple models and show that they are useful for a discussion of the convergence properties of the cmx as well as for the approximate calculation of the autocorrelation function .
in what follows we discuss the solution of the system of @xmath4 equations @xmath5 for the @xmath4 unknowns @xmath6 and @xmath7 , where @xmath8 are known real numbers and @xmath9 an integer .
prony s method is based on the construction of the polynomial@xcite @xmath10 where the polynomial roots @xmath7 , and thereby the polynomial coefficients @xmath11 , are unknown .
it follows from @xmath12 that the coefficients @xmath11 are solutions to the linear system of equations @xmath13 if the matrix @xmath14 with elements @xmath15 , @xmath16 , @xmath17 is nonsingular then the solution is unique .
once we have the polynomial coefficients @xmath11 we obtain its roots @xmath7 and then solve the resulting system of linear equations ( [ eq : main_eqs ] ) ( for example for @xmath18 ) for the remaining unknowns @xmath6 .
the technique just outlined is known since 1795@xcite and is commonly called prony s method@xcite . weiss and
mcdonough@xcite proposed an alternative way of solving the nonlinear equations by means of a partial fraction expansion of pad approximants , and in what follows we describe a different strategy for obtaining the roots @xmath7 developed by amore and fernndez@xcite .
the starting point of this procedure is the homogeneous system of @xmath19 linear equations @xmath20 with @xmath19 unknowns @xmath21 .
there will be nontrivial solutions ( @xmath22 ) only if the determinant vanishes : @xmath23 under such conditions we define @xmath24 so that equations ( [ eq : secular_eqs ] ) become @xmath25 that leads to @xmath26 , @xmath27 .
therefore , it follows from @xmath28 that the roots of the determinant ( [ eq : secular_det ] ) are exactly the roots of the polynomial @xmath29 of prony s method .
we have now at least three alternative procedures for solving the nonlinear equations ( [ eq : main_eqs ] ) .
first , the original prony s method that consists of solving the system of linear equations ( [ eq : linear_pj ] ) for the coefficients of the polynomial @xmath29 , then calculating its roots @xmath7 , and finally solving the resulting system of linear equations ( [ eq : main_eqs ] ) for the remaining unknowns @xmath6 .
second , the partial fraction expansion of the pad approximants proposed by weiss and mcdonough@xcite . third , our approach that consists of obtaining the polynomial @xmath29 from the determinant ( [ eq : secular_det ] ) .
the choice of either of them is probably a matter of taste .
we find our technique quite straightforward and it has revealed a most interesting connection between the exponential expansion of the generating function for the moments and the rayleigh ritz method in the krylov space@xcite . prony s method was developed for fitting a series of exponential functions to given numerical data@xcite . in what follows
we apply it to a completely different problem so that the name prony s method refers only to the strategy for solving the nonlinear equations ( [ eq : main_eqs ] ) .
the general equations developed in the preceding section prove useful for the analysis of the cmx@xcite .
we first consider the moments generating function@xcite @xmath30 where @xmath31 is the hamiltonian operator of the system and @xmath32 is an arbitrary reference state .
the coefficients of its maclaurin series @xmath33 are the moments @xmath34 .
for simplicity we assume that the spectrum of @xmath31 is discrete @xmath35 where @xmath36 , and without loss of generality we choose the eigenvectors of @xmath31 to be orthonormal : @xmath37 . under such conditions
the generating function exhibits the exponential behaviour @xmath38 where the expansion coefficients @xmath39and the energies @xmath40 are unknown .
we can calculate them approximately by means of the method developed in the preceding section and the ansatz @xmath41 where @xmath42 and @xmath43 are the approximations to @xmath44and @xmath40 , respectively .
if we require that the maclaurin expansion of this ansatz yields the first moments @xmath45 exactly we are left with the nonlinear equations ( [ eq : main_eqs ] ) with @xmath46 , @xmath47 and @xmath48 . it is has been proved that matching those expansions at origin is equivalent to the application of the rayleigh
ritz variational method in the krylov space ( rrk)@xcite . note that the determinant ( [ eq : secular_det ] ) becomes the secular determinant of the rrk .
the rrk solutions @xmath49 satisfy @xmath50 and the application of prony s method in the way just outlined yields @xmath51 .
the cmx is based on the generating function @xmath52 that is monotonically decreasing@xcite and the coefficients of its maclaurin series are the connected moments @xmath53 : @xmath54 the recurrence relation@xcite @xmath55 yields the connected moments ( or cummulants ) @xmath56 in terms of the moments @xmath45 .
the main interest in @xmath2 is that it provides a size consistent approach to the ground
state energy@xcite @xmath57 when @xmath58 .
in order to carry out this extrapolation cioslowski@xcite proposed the exponential series ansatz @xmath59 where the unknown parameters @xmath60 are supposed to be real and positive and @xmath61 is the approximation to @xmath62 . matching this expression with the @xmath0expansion ( [ eq : t_exp ] ) leads to the set of equations @xmath63 and @xmath64 we can solve the set of @xmath4 equations ( [ eq : i_j_a_j_b_j ] ) by means of the method of the preceding section with @xmath65 and @xmath66 . in this case
the exponential parameters @xmath60 , @xmath27 are the roots of the pseudo secular determinant @xmath67 once we have them we solve @xmath19 of the resulting @xmath4 linear equations ( [ eq : i_j_a_j_b_j ] ) for the coefficients @xmath42 , @xmath27 , and then we obtain @xmath61 from equation ( [ eq : a0_i1-an ] ) .
cioslowski@xcite and knowles@xcite developed remarkable strategies for obtaining @xmath61 without calculating the other parameters explicitly .
we have just shown that the explicit calculation of all the parameters in the exponential series ansatz ( [ eq : e^(n)(t ) ] ) is quite straightforward .
in order to test the equations developed in the preceding section and show their usefulness in the analysis of the cmx we first consider an exactly solvable model : the dimensionless harmonic oscillator @xmath68 as a trial function we choose one of the examples discussed in earlier papers@xcite @xmath69 for which @xmath70 , @xmath71 .
( the normalization factor is irrelevant to present purposes ) one advantage of this example is that it is not difficult to obtain the generating function exactly @xmath72 note that @xmath73 because @xmath74 in agreement with the discussion in sec .
[ sec : generating ] . by means of the general results of the preceding section we can easily calculate the generating function @xmath2 approximately in two ways : as @xmath75 from equation ( [ eq : z_n(t ) ] ) and directly from equation ( [ eq : e^(n)(t ) ] ) .
fig [ fig : etho2a ] shows that @xmath76 approaches the exact generating function @xmath2 for all @xmath77 as @xmath19 increases . on the other hand , the approximate expression @xmath78 given by eq .
( [ eq : e^(n)(t ) ] ) is an unsatisfactory approximation to the exact @xmath2 as shown in fig .
[ fig : etho2b ] ( except for sufficiently small values of @xmath0 )
. this situation does not appear to improve as @xmath19 increases .
it is found that @xmath79 for @xmath80 because in both cases there is one negative root @xmath60 associated to a negative coefficient @xmath42 .
for example , @xmath81 , @xmath82 , @xmath83 , @xmath84 , @xmath85 , and @xmath86 for @xmath87 . note that the curves in both figures are directly comparable in the sense that the calculation of either @xmath88 and @xmath78 requires exactly the same number of moments @xmath45 ( @xmath89 ) .
the behaviour of @xmath78 for larger @xmath19 may be different ; for example , for @xmath90 there are two negative roots and @xmath91 .
the anomalous behaviour of @xmath78 is due to the poles of @xmath2 in the complex @xmath0plane .
note that eq .
( [ eq : e(t)_ho_exact ] ) exhibits three real poles in the @xmath92plane and the closest to the origin is located at @xmath93 .
it is reasonable to assume that the cmx ansatz @xmath94 should approach the taylor series about @xmath95 for @xmath2 when the approach is successful .
however , this expansion does not converge for @xmath96 which is necessary for matching the @xmath0expansion and exponential series at @xmath1 .
this problem was discussed by amore et al@xcite by means of two level models and we see it here again for the harmonic oscillator . those authors have shown that the cmx approximants for the harmonic oscillator with the trial function ( [ eq : phi(x)_ho ] ) converge towards the second excited state @xmath97 . from the approximate expressions
@xmath94 with @xmath98 we obtain @xmath99 , respectively , in agreement with those results .
this example shows that the cmx approximants may converge to a meaningful result even when the ansatz @xmath78 ( on which the approach is based ) is an unacceptable global approximation to @xmath2 .
knowles@xcite argued that if the connected moments @xmath100 are positive for all @xmath101 then the @xmath102 are real and positive .
later , massano et al@xcite and mancini et al@xcite suggested that knowles argument may not be valid and that it is the hadamard determinants constructed from the @xmath100 that should be positive .
the example just analysed supports the latter conclusion because the first connected moments are all positive ( @xmath103 , @xmath104 , @xmath105 , @xmath106 , @xmath107 , @xmath108 , @xmath109 ) but one of the roots is negative as shown above .
we realize that the general formulas derived in sec .
[ sec : prony ] are most useful for this kind of analysis . in fact , the calculation of the roots @xmath60 by means of present formulas is almost as easy as the calculation of the hadamard determinants and the former provide a much clearer indication of the suitability of the ansatz @xmath78 as an approach for @xmath2 . by means of the example discussed above we do not want to convey the wrong impression that the cmx is suitable for the calculation of excited state energies because it is quite difficult to choose an appropriate trial function for this purpose ( except when we can exploit the symmetry of the problem to make the trial function orthogonal to the ground
state@xcite ) .
the main interest in that exactly solvable problem is to show that we may predict an anomalous behaviour of the cmx by means of the roots @xmath110 .
if they are not real and positive it is reasonable to suspect that the trial function may not be suitable and that it is convenient to choose another one .
for example , if @xmath111 all the roots @xmath60 are positive and the resulting @xmath78 is a remarkably good approximation to @xmath2 for all @xmath112 even at the low orders @xmath113 .
finally , we explore the possibility of calculating of the correlation function @xmath114 where @xmath115 is the initial state normalized to unity and @xmath116 is the state at time @xmath0 .
in particular we concentrate on the real function @xmath117 . as a first example we consider the harmonic oscillator ( [ eq : h_ho ] ) and the trial function @xmath118 the exact correlation function is @xmath119 fig . [
fig : zitho ] shows results for @xmath120 .
there is no doubt that @xmath121 converges towards @xmath117 as @xmath19 increases .
the approximation is satisfactory within the first period but the errors accumulate as @xmath0 increases .
this fact is not surprising because the approximate expression is constructed from the maclaurin series for @xmath122 .
the second example is the anharmonic oscillator @xmath123 and the same trial function ( [ eq : phi(0)_ho ] ) .
[ fig : zitaho ] shows that the approximate results for @xmath120 follow the same trend as in the case of the harmonic oscillator . at first sight
it may be surprising that the convergence rate for the anharmonic oscillator appears to be greater than for the harmonic one .
the reason is that the chosen trial state exhibits a greater overlap with the ground state of the former .
note that the magnitude of the coefficient @xmath61 of @xmath124 tells us that @xmath125 is approximately 0.943 for the harmonic oscillator and 0.981 for the anharmonic one .
as a two dimensional example we consider the anharmonic oscillator @xmath126 where @xmath127 , and the trial function @xmath128 fig .
[ fig : zitpe ] shows results for @xmath129 that look quite similar to those for the one dimensional models .
one does not expect that it may be possible to construct @xmath130 from its taylor expansion about @xmath1 in all the cases .
the problems discussed above are simple examples of single
well oscillators . in the case of the double well oscillator @xmath131 and the trial state localized in one of the wells
@xmath132 the approximants to @xmath117 do not appear to converge although the results for @xmath133 look reasonable .
if , on the other hand , we locate the initial state on top of the barier ( for example , eq .
( [ eq : phi(0)_ho ] ) ) then the results are as accurate as those for the single wells .
the main goal of this paper is to show that prony s method provides the full solution of the cmx equations in a straightforward and simple way .
we have outlined the connection between prony s method and the procedure developed recently by amore and fernndez@xcite .
although both approaches are equivalent and prony s one is known since long ago it seems that the pseudo secular determinants obtained by amore and fernndez are not available elsewhere . in spite of the fact that the harmonic oscillator with the trial function ( [ eq : phi(x)_ho ] ) was chosen as an illustrative example in two earlier papers@xcite we think that present discussion based on the exact generating function ( [ eq : e(t)_ho_exact ] ) is clearer and provides more information about the behaviour of the cmx ansatz @xmath78 .
in particular , we have shown that this simple problem clearly illustrates that knowles conclusion about the sign of the connected moments@xcite is not valid as argued before by massano et al@xcite and mancini et al@xcite .
it is clear that the roots @xmath102 provide a clear indication of the success of the cmx and the method developed by amore and fernndez@xcite , as well as prony s method , make their calculation straightforward .
both procedures may be applied to the calculation of the autocorrelation function for simple quantum mechanical models .
if the approach converges then one obtains a reasonable analytic expression for the autocorrelation function within the first period of oscillation . for the harmonic oscillator ( line ) and
approximants @xmath134 with @xmath135 ( squares , blue ) , @xmath87 ( filled squares , red ) , @xmath136 ( circles , green ) and @xmath137 ( filled circle , black).,width=340 ] | we show that prony s method provides the full solution to the nonlinear equations of the connected moments expansion ( cmx ) .
knowledge of all the parameters in the cmx ansatz is useful for the analysis of the convergence properties of the approach .
prony s method is also suitable for the calculation of the correlation function for simple quantum mechanical models . |
title of program : siliconiap computer hardware and operating system : any shared memory computer running under unix or linux programming language : fortran90 with openmp compiler directives memory requirements : roughly 150 words per atom no . of bits in a word : 64 no . of processors
used : tested on up to 4 processors has the code been vectorized or parallelized : parallelized withe openmp no .
of bytes in distributed program , including test data , etc : 50 000 distribution format : compressed tar file keywords : silicon , interatomic potential , force field , molecular dynamics nature of physical problem : condensed matter physics method of solution : interatomic potential restrictions on the complexity of the problem : none typical running time : 30 @xmath0sec per step and per atom on a compaq dec alpha unusual features of the program : none
due to its technological importance , silicon is one of the most studied materials . for small
system sizes ab - initio density functional calculations @xcite are the preferred approach .
unfortunately this kind of calculation becomes unfeasible for larger systems required to study problems such as interfaces or extended defects . for this type of calculations
one resorts to force fields which are several orders of magnitude faster .
recent progress in the development of force fields has demonstrated that they can be a reliable tool for such studies .
a highly accurate silicon force field has been developed by lenosky and coworkers @xcite .
its transferability has been demonstrated by extensive tests containing both bulk and cluster systems @xcite .
its accuracy is in part due to the fact that second nearest neighbor interactions are included .
this makes it unfortunately somewhat slower than force fields containing only nearest neighbor interactions . in the following a highly optimized parallel implementation of this force field
will be presented that allows large scale calculations with this force field .
the parallelization is achieved by using openmp , an emerging industry standard for medium size shared memory parallel computers .
molecular dynamics calculations @xcite have also been parallelized on distributed memory supercomputers @xcite .
this approach is considerably more complex than the one presented here . since few researches
have access to massively parallel supercomputers and are willing to overcome the complexities of doing molecular dynamics on such machines , medium scale parallelization @xcite of molecular dynamics has an important place in practice .
user friendliness was one of the major design goals in the development of this routine . using fortran90 made it possible to hide all the complexities in an object oriented fashion from the user .
the calling sequence is just .... call lenosky(nat , alat , rxyz , fxyz , ener , coord , ener_var , coord_var , count ) .... on input the user has to specify the number of atoms , @xmath1 , the vector @xmath2 containing the 3 lattice constant of the orthorhombic periodic volume and the atomic positions @xmath3 .
the program then returns the total energy , @xmath4 , the forces , @xmath5 , the average coordination number , the variation of the energy per atom and of the coordination number as well as an counter that is increased in each call .
in particular the user has not to supply any verlet list .
since the calculation of the forces is typically much more expensive than the update of the atomic positions in molecular dynamics or geometry optimizations , we expect that the subroutine will be called in most cases from within a serial program . in case
the user is on a shared memory machine the subroutine will then nevertheless be executed in parallel if the program is compiled with the appropriate openmp options .
in addition the subroutine can of course also be used on a serial machine . in this case
all the parallelization directives are considered by the compiler to be comments .
the verlet list gives all the atoms that are contained within the potential cutoff distance @xmath6 of any given atom .
typically the verlet list consists of two integer arrays .
the first array , called @xmath7 in this work , points to the first / last neighbor position in the second array @xmath8 that contains the numbering of the atoms that are neighbors .
a straightforward implementation for a non - periodic system containing @xmath1 atoms is shown below . in this simple case
the search through all atoms is sequential with respect to their numbering and it is redundant to give both the starting positions @xmath9 and the ending position @xmath10 , since @xmath11 .
but in the more complicated linear scaling algorithm to be presented below , both will be needed . ....
indc=0 do 10 iat=1,nat c starting position lsta(1,iat)=indc+1 do 20 jat=1,nat if ( jat.ne.iat ) then xrel1= rxyz(1,jat)-rxyz(1,iat ) xrel2= rxyz(2,jat)-rxyz(2,iat ) xrel3= rxyz(3,jat)-rxyz(3,iat ) rr2=xrel1**2 + xrel2**2 + xrel3**2 if ( rr2 .le .
cut**2 ) then indc = indc+1 c nearest neighbor numbers lstb(indc)=jat endif endif 20 continue c ending position lsta(2,iat)=indc 10 continue .... this straightforward implementations has a quadratic scaling with respect to the numbers of atoms . due to this
scaling the calculation of the verlet list starts to dominate the linear scaling calculation of the energies and forces for system sizes of more than 10 000 atoms .
it is therefore good practice to calculate the verlet list with a modified algorithm that has linear scaling @xcite as well .
to do this one first subdivides the system into boxes that have a side length that is equal to or larger than @xmath6 and then finds all the atoms that are contained in each box .
the cpu time for this first step is less than 1 percent of the entire verlet list calculation .
hence this part was not parallelized .
it could significantly affect the parallel performance according to amdahls law @xcite only if more than 50 processors are used .
the largest smp machines at our disposal had however only 4 processors . to implement periodic boundary conditions all atoms within a distance @xmath6 of the boundary of the periodic volume
are replicated on the opposite part as shown in figure [ period ] .
this part is equally well less than 1 percent of the cpu time for the verlet list for a 8000 atom system .
being a surface term it becomes even smaller for larger systems .
consequently it was nt parallelized either .
( -1.5,0.5 ) after these two preparing steps one has to search only among all the atoms in the reference cell containing the atom for which one wants to find its neighbors as well as all the atoms in the cells neighboring this reference cell ( 26 cells in 3 dimensions ) .
this implies that starting and ending positions @xmath7 for the atoms @xmath12 to @xmath1 are not calculated in a sequential way , necessitating , as mentioned before , separate starting and ending positions in the array @xmath7 .
the corresponding parallel code is shown below .
the indices @xmath13 refer to the cells , @xmath14 contains the number of atoms in cell @xmath13 and @xmath15 their numbering .
the array @xmath16 saves the relative positions and distances that will again be needed in the loop calculating the forces and energies .
each thread has its own starting position @xmath17 in the shared memory space @xmath8 and these starting positions are uniformly distributed .
this approach allows the different threads to work independently .
the resulting speedup is much higher than the one that one would obtain by calculating in the parallel version an array @xmath8 that is identical to the one from the serial version .
if there are on the average more neighbors than expected ( 24 by default ) the allocated space becomes too small . in this case
the array @xmath8 is deallocated and a new larger version is allocated .
this check for sufficient memory requires some minimal amount of coordination among the processors and is implemented by a critical section .
if a reallocation is necessary , a message is written into an file to alert the user of the inefficiency due to the need of a second calculation of the verlet list . ....
allocate(lsta(2,nat ) ) nnbrx=24 2345 nnbrx=3*nnbrx/2 allocate(lstb(nnbrx*nat),rel(5,nnbrx*nat ) ) indlstx=0 !
$ omp parallel & !
$ omp private(iat , cut2,iam , ii , indlst , l1,l2,l3,myspace , npr ) & ! $ omp shared ( indlstx , nat , nn , nnbrx , ncx , ll1,ll2,ll3,icell , lsta , lstb , lay , & !
$ omp rel , rxyz , cut , myspaceout ) npr=1 ! $ npr = omp_get_num_threads ( ) iam=0 !
$ iam = omp_get_thread_num ( ) cut2=cut**2 myspace=(nat*nnbrx)/npr if ( iam.eq.0 ) myspaceout = myspace !
verlet list , relative positions indlst=0 do 6000,l3=0,ll3 - 1 do 6000,l2=0,ll2 - 1 do 6000,l1=0,ll1 - 1 do 6600,ii=1,icell(0,l1,l2,l3 ) iat = icell(ii , l1,l2,l3 ) if ( ( ( iat-1)*npr)/nat .eq .
iam ) then lsta(1,iat)=iam*myspace+indlst+1 call sublstiat(iat , nn , ncx , ll1,ll2,ll3,l1,l2,l3,myspace , & rxyz , icell , lstb(iam*myspace+1),lay , rel(1,iam*myspace+1),cut2,indlst ) lsta(2,iat)=iam*myspace+indlst endif 6600 continue 6000 continue !
$ omp critical indlstx = max(indlstx , indlst ) !
$ omp end critical !
$ omp end parallel if ( indlstx.gt.myspaceout ) then write(10 , * ) count,'nnbrx too small ' , nnbrx deallocate(lstb , rel ) goto 2345 endif subroutine sublstiat(iat , nn , ncx , ll1,ll2,ll3,l1,l2,l3,myspace , & rxyz , icell , lstb , lay , rel , cut2,indlst ) implicit real*8 ( a - h , o - z ) dimension rxyz(3,nn),lay(nn),icell(0:ncx,-1:ll1,-1:ll2,-1:ll3 ) , & lstb(0:myspace-1),rel(5,0:myspace-1 ) do 6363,k3=l3 - 1,l3 + 1 do 6363,k2=l2 - 1,l2 + 1 do 6363,k1=l1 - 1,l1 + 1 do 6363,jj=1,icell(0,k1,k2,k3 ) jat = icell(jj ,
k1,k2,k3 ) if ( jat.eq.iat ) goto 6363 xrel= rxyz(1,iat)-rxyz(1,jat ) yrel=
rxyz(2,iat)-rxyz(2,jat ) zrel= rxyz(3,iat)-rxyz(3,jat ) rr2=xrel**2 + yrel**2 + zrel**2 if ( rr2 .le .
cut2 ) then indlst = min(indlst , myspace-1 ) lstb(indlst)=lay(jat ) tt = sqrt(rr2 ) tti=1.d0/tt rel(1,indlst)=xrel*tti rel(2,indlst)=yrel*tti rel(3,indlst)=zrel*tti rel(4,indlst)=tt rel(5,indlst)=tti indlst= indlst+1 endif 6363 continue return end ....
the computationally most important part taking some 80 percent of the cpu time is the calculation of the energies and forces .
the energy expression for the lenosky force field is given by @xmath18 all the functions ( @xmath19 , @xmath20 , @xmath21 , @xmath22 , @xmath23 ) in this energy expression are given by cubic splines .
the subroutine for evaluating the cubic spline is listed below .
the case that the argument is outside the cubic spline interval @xmath24 $ ] is rare and unimportant for performance considerations .
the important cubic spline case is characterized by many dependencies . in the case of such dependencies
the latency of the functional unit pipeline comes into play and reduces the attainable speed @xcite .
a latency of some 20 cycles comes from the first two statements ( tt=(x - tmin)*hi ; klo = tt ) alone , requiring arithmetic operations and a floating point to integer conversion . for this reason
the calculation of tt was taken out of the ( most likely occurring ) else block to overlap its evaluation with the evaluation of the if clauses . to further speed up the evaluation of the splines the structure of the energy expression [ energy ]
was exploited . in the computationally most important loop over @xmath25 and @xmath26 two splines ( @xmath27 and @xmath28 )
have to be evaluated .
inlining by hand the subroutine splint for both evaluations and calculating alternatingly one step of the first spline evaluation and one step of the second spline evaluation introduces two independent streams .
this reduces the effect of latencies and boosts speed .
compilers are not able to do these complex type of optimizations .
the best performance after these optimizations was obtained with low level compiler optimization flags ( -o3 -qarch = pwr3 -qtune - pwr3 on ibm power3 , -o2 on the compaq dec alpha , -o2 -xw on intel pentium4 ) .... subroutine splint(ya , y2a , tmin , tmax , hsixth , h2sixth , hi ,
n , x , y , yp ) implicit real*8 ( a - h , o - z ) dimension y2a(0:n-1),ya(0:n-1 ) ! interpolate if the argument is outside the cubic spline interval [ tmin , tmax ] tt=(x - tmin)*hi if ( x.lt.tmin ) then yp = hi*(ya(1)-ya(0 ) ) - & ( y2a(1)+2.d0*y2a(0 ) ) * hsixth y = ya(0 ) + ( x - tmin)*yp else if ( x.gt.tmax ) then yp = hi*(ya(n-1)-ya(n-2 ) ) + & ( 2.d0*y2a(n-1)+y2a(n-2 ) ) * hsixth y = ya(n-1 ) + ( x - tmax)*yp
! otherwise evaluate cubic spline else klo = tt khi = klo+1 ya_klo = ya(klo ) y2a_klo = y2a(klo ) b = tt - klo a=1.d0-b ya_khi = ya(khi ) y2a_khi = y2a(khi ) b2=b*b y = a*ya_klo
yp = ya_khi - ya_klo a2=a*a cof1=a2 - 1.d0 cof2=b2 - 1.d0 y = y+b*ya_khi yp = hi*yp cof3=3.d0*b2 cof4=3.d0*a2 cof1=a*cof1 cof2=b*cof2 cof3=cof3 - 1.d0 cof4=cof4 - 1.d0 yt1=cof1*y2a_klo yt2=cof2*y2a_khi ypt1=cof3*y2a_khi ypt2=cof4*y2a_klo y = y + ( yt1+yt2)*h2sixth yp = yp + ( ypt1 - ypt2 ) * hsixth endif return end .... the final single processor performance for the entire subroutine is 460 mflops on a compaq dec alpha at 833 mhz , 300 mflops on a ibm power3 at 350 mhz and 550 mflops on a pentium 4 . in order to obtain a high parallel speedup in this central part of the subroutine
the threads are completely decoupled .
this was done by introducing private copies for each thread to accumulate the energies @xmath29 and forces @xmath30 .
the global energy and force are summed up in an additional loop at the end of the parallel region in a critical section . .... ! $ omp parallel & !
$ omp private(iam , npr , iat , iat1,iat2,lot , istop , tcoord , tcoord2 , & ! $ omp tener , tener2,txyz , f2ij , f3ij , f3ik , npjx , npjkx ) & ! $ omp shared ( nat , nnbrx , lsta , lstb , rel , ener , ener2,fxyz , coord , coord2,istopg ) npr=1 ! $ npr = omp_get_num_threads ( ) iam=0
! $ iam = omp_get_thread_num ( ) npjx=300 ; npjkx=3000 istopg=0 if ( npr.ne.1 ) then ! parallel case !
create temporary private scalars for reduction sum on energies and ! temporary private array for reduction sum on forces !
$ omp critical allocate(txyz(3,nat),f2ij(3,npjx),f3ij(3,npjkx),f3ik(3,npjkx ) ) ! $ omp
end critical if ( iam.eq.0 ) then ener=0.d0 ener2=0.d0 coord=0.d0 coord2=0.d0 do 121,iat=1,nat fxyz(1,iat)=0.d0 fxyz(2,iat)=0.d0 121 fxyz(3,iat)=0.d0 endif lot = nat / npr+.999999999999d0 iat1=iam*lot+1 iat2=min((iam+1)*lot , nat ) call subfeniat(iat1,iat2,nat , lsta , lstb , rel , tener , tener2 , & tcoord , tcoord2,nnbrx , txyz , f2ij , npjx , f3ij , npjkx , f3ik , istop ) !
$ omp critical ener = ener+tener ener2=ener2+tener2 coord = coord+tcoord coord2=coord2+tcoord2 istopg = istopg+istop do 8093,iat=1,nat fxyz(1,iat)=fxyz(1,iat)+txyz(1,iat ) fxyz(2,iat)=fxyz(2,iat)+txyz(2,iat ) fxyz(3,iat)=fxyz(3,iat)+txyz(3,iat ) 8093 continue !
$ omp end critical deallocate(txyz , f2ij , f3ij , f3ik ) else !
serial case iat1=1 iat2=nat allocate(f2ij(3,npjx),f3ij(3,npjkx),f3ik(3,npjkx ) ) call subfeniat(iat1,iat2,nat , lsta , lstb , rel , ener , ener2 , & coord , coord2,nnbrx , fxyz , f2ij , npjx , f3ij , npjkx , f3ik , istop ) deallocate(f2ij , f3ij , f3ik ) endif !
$ omp end parallel if ( istopg.gt.0 ) stop ' dimension error ( see warning above ) ' ener_var = ener2/nat-(ener / nat)**2 coord = coord / nat coord_var = coord2/nat - coord**2 deallocate(rxyz , icell , lay , lsta , lstb , rel ) end subroutine subfeniat(iat1,iat2,nat , lsta , lstb , rel , tener , tener2 , & tcoord , tcoord2,nnbrx , txyz , f2ij , npjx , f3ij , npjkx , f3ik , istop ) implicit real*8 ( a - h , o - z ) dimension lsta(2,nat),lstb(nnbrx*nat),rel(5,nnbrx*nat),txyz(3,nat ) dimension
f2ij(3,npjx),f3ij(3,npjkx),f3ik(3,npjkx ) initialize data ........ ! create temporary private scalars for reduction sum on energies and tener=0.d0 tener2=0.d0 tcoord=0.d0 tcoord2=0.d0 istop=0 do 121,iat=1,nat txyz(1,iat)=0.d0 txyz(2,iat)=0.d0 121 txyz(3,iat)=0.d0 !
calculation of forces , energy do 1000,iat = iat1,iat2 dens2=0.d0 dens3=0.d0 jcnt=0 jkcnt=0 coord_iat=0.d0 ener_iat=0.d0 do 2000,jbr = lsta(1,iat),lsta(2,iat ) jat = lstb(jbr ) jcnt = jcnt+1 if ( jcnt.gt.npjx ) then write(6 , * ) ' warning : enlarge npjx ' istop=1 endif fxij = rel(1,jbr ) fyij = rel(2,jbr ) fzij = rel(3,jbr ) rij = rel(4,jbr ) sij = rel(5,jbr ) ! coordination number calculated with soft cutoff between first and !
second nearest neighbor if ( rij.le.2.36d0 ) then coord_iat = coord_iat+1.d0 else if ( rij.ge.3.83d0 ) then else x=(rij-2.36d0)*(1.d0/(3.83d0 - 2.36d0 ) ) coord_iat = coord_iat+(2*x+1.d0)*(x-1.d0)**2 endif !
pairpotential term call splint(cof_phi , dof_phi , tmin_phi
, tmax_phi , & hsixth_phi , h2sixth_phi , hi_phi,10,rij , e_phi , ep_phi ) ener_iat = ener_iat+(e_phi*.5d0 ) txyz(1,iat)=txyz(1,iat)-fxij*(ep_phi*.5d0 ) txyz(2,iat)=txyz(2,iat)-fyij*(ep_phi*.5d0 ) txyz(3,iat)=txyz(3,iat)-fzij*(ep_phi*.5d0 ) txyz(1,jat)=txyz(1,jat)+fxij*(ep_phi*.5d0 ) txyz(2,jat)=txyz(2,jat)+fyij*(ep_phi*.5d0 ) txyz(3,jat)=txyz(3,jat)+fzij*(ep_phi*.5d0 ) ! 2 body embedding term call splint(cof_rho , dof_rho , tmin_rho , tmax_rho , & hsixth_rho , h2sixth_rho , hi_rho,11,rij , rho , rhop ) dens2=dens2+rho f2ij(1,jcnt)=fxij*rhop f2ij(2,jcnt)=fyij*rhop f2ij(3,jcnt)=fzij*rhop ! 3 body embedding term call splint(cof_fff , dof_fff , tmin_fff , tmax_fff , & hsixth_fff , h2sixth_fff , hi_fff,10,rij , fij , fijp ) do 3000,kbr = lsta(1,iat),lsta(2,iat ) kat = lstb(kbr ) if ( kat.lt.jat ) then jkcnt = jkcnt+1 if ( jkcnt.gt.npjkx ) then write(6 , * ) ' warning : enlarge npjkx ' istop=1 endif ! begin optimized version rik = rel(4,kbr ) if ( rik.gt.tmax_fff ) then fikp=0.d0 ; fik=0.d0 gjik=0.d0 ; gjikp=0.d0 ; sik=0.d0 costheta=0.d0 ; fxik=0.d0 ; fyik=0.d0 ; fzik=0.d0 else if ( rik.lt.tmin_fff ) then fxik = rel(1,kbr ) fyik = rel(2,kbr ) fzik = rel(3,kbr ) costheta = fxij*fxik+fyij*fyik+fzij*fzik sik = rel(5,kbr ) fikp = hi_fff*(cof_fff(1)-cof_fff(0 ) ) - & ( dof_fff(1)+2.d0*dof_fff(0 ) ) * hsixth_fff fik = cof_fff(0 ) + ( rik - tmin_fff)*fikp tt_ggg=(costheta - tmin_ggg)*hi_ggg if ( costheta.gt.tmax_ggg ) then gjikp = hi_ggg*(cof_ggg(8 - 1)-cof_ggg(8 - 2 ) ) + & ( 2.d0*dof_ggg(8 - 1)+dof_ggg(8 - 2 ) ) * hsixth_ggg gjik = cof_ggg(8 - 1 ) + ( costheta - tmax_ggg)*gjikp else klo_ggg = tt_ggg khi_ggg = klo_ggg+1 cof_ggg_klo = cof_ggg(klo_ggg ) dof_ggg_klo = dof_ggg(klo_ggg )
b_ggg = tt_ggg - klo_ggg a_ggg=1.d0-b_ggg cof_ggg_khi = cof_ggg(khi_ggg ) dof_ggg_khi = dof_ggg(khi_ggg ) b2_ggg = b_ggg*b_ggg gjik = a_ggg*cof_ggg_klo gjikp = cof_ggg_khi - cof_ggg_klo a2_ggg = a_ggg*a_ggg cof1_ggg = a2_ggg-1.d0 cof2_ggg = b2_ggg-1.d0 gjik = gjik+b_ggg*cof_ggg_khi
gjikp = hi_ggg*gjikp cof3_ggg=3.d0*b2_ggg cof4_ggg=3.d0*a2_ggg cof1_ggg = a_ggg*cof1_ggg cof2_ggg = b_ggg*cof2_ggg cof3_ggg = cof3_ggg-1.d0 cof4_ggg = cof4_ggg-1.d0 yt1_ggg = cof1_ggg*dof_ggg_klo yt2_ggg = cof2_ggg*dof_ggg_khi ypt1_ggg = cof3_ggg*dof_ggg_khi ypt2_ggg = cof4_ggg*dof_ggg_klo gjik = gjik + ( yt1_ggg+yt2_ggg)*h2sixth_ggg gjikp = gjikp + ( ypt1_ggg - ypt2_ggg ) * hsixth_ggg endif else fxik = rel(1,kbr ) tt_fff = rik - tmin_fff costheta = fxij*fxik fyik = rel(2,kbr ) tt_fff = tt_fff*hi_fff costheta = costheta+fyij*fyik fzik = rel(3,kbr ) klo_fff = tt_fff costheta = costheta+fzij*fzik sik = rel(5,kbr ) tt_ggg=(costheta - tmin_ggg)*hi_ggg if ( costheta.gt.tmax_ggg ) then gjikp = hi_ggg*(cof_ggg(8 - 1)-cof_ggg(8 - 2 ) ) + & ( 2.d0*dof_ggg(8 - 1)+dof_ggg(8 - 2 ) ) * hsixth_ggg gjik = cof_ggg(8 - 1 ) + ( costheta - tmax_ggg)*gjikp khi_fff = klo_fff+1 cof_fff_klo = cof_fff(klo_fff ) dof_fff_klo = dof_fff(klo_fff ) b_fff = tt_fff - klo_fff a_fff=1.d0-b_fff cof_fff_khi = cof_fff(khi_fff ) dof_fff_khi = dof_fff(khi_fff ) b2_fff = b_fff*b_fff fik = a_fff*cof_fff_klo fikp = cof_fff_khi - cof_fff_klo a2_fff = a_fff*a_fff cof1_fff = a2_fff-1.d0 cof2_fff = b2_fff-1.d0 fik = fik+b_fff*cof_fff_khi fikp = hi_fff*fikp cof3_fff=3.d0*b2_fff cof4_fff=3.d0*a2_fff cof1_fff = a_fff*cof1_fff cof2_fff = b_fff*cof2_fff cof3_fff = cof3_fff-1.d0 cof4_fff = cof4_fff-1.d0 yt1_fff = cof1_fff*dof_fff_klo yt2_fff = cof2_fff*dof_fff_khi ypt1_fff = cof3_fff*dof_fff_khi ypt2_fff = cof4_fff*dof_fff_klo fik = fik + ( yt1_fff+yt2_fff)*h2sixth_fff fikp = fikp + ( ypt1_fff - ypt2_fff ) * hsixth_fff else klo_ggg = tt_ggg khi_ggg = klo_ggg+1 khi_fff = klo_fff+1 cof_ggg_klo = cof_ggg(klo_ggg ) cof_fff_klo = cof_fff(klo_fff ) dof_ggg_klo = dof_ggg(klo_ggg ) dof_fff_klo = dof_fff(klo_fff ) b_ggg = tt_ggg - klo_ggg b_fff = tt_fff - klo_fff a_ggg=1.d0-b_ggg a_fff=1.d0-b_fff cof_ggg_khi = cof_ggg(khi_ggg ) cof_fff_khi = cof_fff(khi_fff ) dof_ggg_khi = dof_ggg(khi_ggg ) dof_fff_khi = dof_fff(khi_fff ) b2_ggg = b_ggg*b_ggg b2_fff = b_fff*b_fff gjik = a_ggg*cof_ggg_klo fik = a_fff*cof_fff_klo gjikp = cof_ggg_khi - cof_ggg_klo fikp = cof_fff_khi - cof_fff_klo a2_ggg = a_ggg*a_ggg a2_fff = a_fff*a_fff cof1_ggg = a2_ggg-1.d0 cof1_fff = a2_fff-1.d0 cof2_ggg = b2_ggg-1.d0 cof2_fff = b2_fff-1.d0 gjik = gjik+b_ggg*cof_ggg_khi fik = fik+b_fff*cof_fff_khi gjikp = hi_ggg*gjikp fikp = hi_fff*fikp cof3_ggg=3.d0*b2_ggg cof3_fff=3.d0*b2_fff cof4_ggg=3.d0*a2_ggg cof4_fff=3.d0*a2_fff cof1_ggg = a_ggg*cof1_ggg cof1_fff = a_fff*cof1_fff cof2_ggg = b_ggg*cof2_ggg cof2_fff = b_fff*cof2_fff cof3_ggg = cof3_ggg-1.d0 cof3_fff = cof3_fff-1.d0 cof4_ggg = cof4_ggg-1.d0 cof4_fff = cof4_fff-1.d0 yt1_ggg = cof1_ggg*dof_ggg_klo yt1_fff = cof1_fff*dof_fff_klo yt2_ggg = cof2_ggg*dof_ggg_khi yt2_fff = cof2_fff*dof_fff_khi ypt1_ggg = cof3_ggg*dof_ggg_khi ypt1_fff = cof3_fff*dof_fff_khi ypt2_ggg = cof4_ggg*dof_ggg_klo ypt2_fff = cof4_fff*dof_fff_klo gjik = gjik + ( yt1_ggg+yt2_ggg)*h2sixth_ggg fik = fik + ( yt1_fff+yt2_fff)*h2sixth_fff gjikp = gjikp + ( ypt1_ggg - ypt2_ggg ) * hsixth_ggg fikp = fikp + ( ypt1_fff - ypt2_fff ) * hsixth_fff endif endif !
end optimized version tt = fij*fik dens3=dens3+tt*gjik t1=fijp*fik*gjik t2=sij*(tt*gjikp ) f3ij(1,jkcnt)=fxij*t1 + ( fxik - fxij*costheta)*t2 f3ij(2,jkcnt)=fyij*t1 + ( fyik - fyij*costheta)*t2 f3ij(3,jkcnt)=fzij*t1 + ( fzik - fzij*costheta)*t2 t3=fikp*fij*gjik t4=sik*(tt*gjikp ) f3ik(1,jkcnt)=fxik*t3 + ( fxij - fxik*costheta)*t4 f3ik(2,jkcnt)=fyik*t3 + ( fyij - fyik*costheta)*t4 f3ik(3,jkcnt)=fzik*t3 + ( fzij - fzik*costheta)*t4 endif 3000 continue 2000 continue dens = dens2+dens3 call splint(cof_uuu , dof_uuu , tmin_uuu , tmax_uuu , & hsixth_uuu , h2sixth_uuu , hi_uuu,8,dens , e_uuu , ep_uuu ) ener_iat = ener_iat+e_uuu ! only now ep_uu is known and the forces can be calculated , lets loop again jcnt=0 jkcnt=0 do 2200,jbr = lsta(1,iat),lsta(2,iat ) jat = lstb(jbr ) jcnt = jcnt+1 txyz(1,iat)=txyz(1,iat)-ep_uuu*f2ij(1,jcnt ) txyz(2,iat)=txyz(2,iat)-ep_uuu*f2ij(2,jcnt ) txyz(3,iat)=txyz(3,iat)-ep_uuu*f2ij(3,jcnt ) txyz(1,jat)=txyz(1,jat)+ep_uuu*f2ij(1,jcnt ) txyz(2,jat)=txyz(2,jat)+ep_uuu*f2ij(2,jcnt ) txyz(3,jat)=txyz(3,jat)+ep_uuu*f2ij(3,jcnt ) !
3 body embedding term do 3300,kbr = lsta(1,iat),lsta(2,iat ) kat = lstb(kbr ) if ( kat.lt.jat ) then jkcnt = jkcnt+1 txyz(1,iat)=txyz(1,iat)-ep_uuu*(f3ij(1,jkcnt)+f3ik(1,jkcnt ) ) txyz(2,iat)=txyz(2,iat)-ep_uuu*(f3ij(2,jkcnt)+f3ik(2,jkcnt ) ) txyz(3,iat)=txyz(3,iat)-ep_uuu*(f3ij(3,jkcnt)+f3ik(3,jkcnt ) ) txyz(1,jat)=txyz(1,jat)+ep_uuu*f3ij(1,jkcnt ) txyz(2,jat)=txyz(2,jat)+ep_uuu*f3ij(2,jkcnt ) txyz(3,jat)=txyz(3,jat)+ep_uuu*f3ij(3,jkcnt ) txyz(1,kat)=txyz(1,kat)+ep_uuu*f3ik(1,jkcnt ) txyz(2,kat)=txyz(2,kat)+ep_uuu*f3ik(2,jkcnt ) txyz(3,kat)=txyz(3,kat)+ep_uuu*f3ik(3,jkcnt ) endif 3300 continue 2200 continue tener = tener+ener_iat tener2=tener2+ener_iat**2 tcoord = tcoord+coord_iat tcoord2=tcoord2+coord_iat**2 1000 continue return end .... in addition to the energy and the forces the program still returns the coordination number as well as the variance of the energy per atom and the coordination number .
the coordination number is calculated using a soft cutoff between the first and second nearest neighbor distance .
these extra calculations are very cheap and not visible as an increase in the cpu time
table [ speedup_dec ] shows the final overall speedups obtained by the program .
the results were obtained for an 8000 atom system , but the cpu time per call and atom is nearly independent of system size . .
timings in @xmath0sec for a combined evaluation of the forces and the energy per particle as well as the corresponding speedups ( in parentheses ) on an ibm sp3 based on a 375 mhz power3 processor , on a compaq sc 232 based on a 833mhz ev67 processor and on an intel pentium4 biprocessor at 2 ghz [ speedup_dec ] [ cols="<,^,^,^",options="header " , ] obtaining such high speedups was not straightforward .
only the compaq fortran90 compiler was able to use in the original version of the program the openmp parallel do directive to obtain a good speedup .
both the ibm compiler and the intel compiler failed . in order to get the performances of table [ speedup_dec ] , it was necessary to encapsulate the workload of the different threads into the subroutines sublstias and subfen , which amounts to doing the parallelization quasi by hand . using allocatable arrays in connection with openmp turned also out to be tricky . because of these problems , the parallelization was much more painful that one might expect for a shared memory model .
the results show that simulations for very large silicon systems are feasible on relatively cheap serial or parallel computers accessible to a large number of researches . | the force field by lenosky and coworkers is the latest force field for silicon which is one of the most studied materials .
it has turned out to be highly accurate in a large range of test cases .
the optimization and parallelization of this force field using openmp and fortan90 is described here .
the optimized program allows us to handle a very large number of silicon atoms in large scale simulations .
since all the parallelization is hidden in a single subroutine that returns the total energies and forces , this subroutine can be called from within a serial program in an user friendly way .
-0.75 in |
hard exclusive meson production in deep inelastic lepton scattering provides access to the unknown generalized parton distributions ( gpds ) of the nucleon @xcite . for such reactions
, it has been shown that for longitudinal virtual photons , the @xmath3 amplitude can be factorized into a hard lepton - scattering part and two soft parts which parameterize the structure of the nucleon by gpds and the structure of the produced meson by distribution amplitude @xcite .
gpds reflect the 3-dimensional structure of the nucleon and contain information about the total angular momentum carried by partons in the nucleon .
hard exclusive production of @xmath0 mesons is sensitive to the gpds @xmath4 and @xmath5 which are the ones related to the total angular momenta @xmath6 and @xmath7 of quarks and gluons in the nucleon @xcite .
the gpd @xmath4 is already somewhat constrained , while the gpd @xmath5 is still unknown . in the case of a transversely polarized target ,
the interference between the gpds @xmath4 and @xmath5 was shown to lead to a transverse target - spin asymmetry ( ttsa ) @xcite .
in contrast to the cross section measurements , the ttsa depends linearly on the helicity - flip distribution @xmath5 with no kinematic suppression of its contribution with respect to the other gpds . therefore the ttsa of exclusive @xmath0 production can constrain the total angular momenta @xmath6 and @xmath7 .
for an unpolarized ( u ) beam and a transversely ( t ) polarized target the ttsa @xmath8 is defined as @xmath9 where the target polarization @xmath10 is defined w.r.t . the lepton beam direction and
the angles @xmath11 and @xmath12 are the azimuthal angles of , respectively , the produced @xmath0 meson and the target spin vector around the virtual photon direction w.r.t . the lepton scattering plane ( see figure [ fig : angle ] ) @xcite .
the cross section of exclusive @xmath0 production can be factorized in terms of angular dependent and angle - independent parts : @xmath13 where @xmath14 is the bjorken scaling variable , @xmath15 is the squared virtual - photon four - momentum , @xmath16 . here
@xmath17 is the squared four - momentum transfer to the target and @xmath18 represents the minimum value of @xmath17 .
the complete expression for the cross section of @xmath0 production is given in @xcite .
the angular distribution @xmath19 can be written , @xmath15 and @xmath20 are omitted . ] in terms of asymmetries : @xmath21 where @xmath22 is the unpolarized asymmetry with @xmath23 , @xmath24 being the unpolarized angular distributions and @xmath25 is the transverse asymmetry with the transversely polarized angular distribution @xmath26 .
r0.45 since the factorization theorem is proven for longitudinal photons only @xcite , the asymmetry of @xmath0 mesons induced from longitudinal photons is of theoretical interest . under the assumption of @xmath2-channel helicity conservation ( schc ) , which implies that a longitudinal vector meson originates from a longitudinal photon ,
the longitudinal component of the asymmetry is obtained experimentally through the decay angular distribution of @xmath0 ( @xmath27 ) .
each @xmath0 helicity state ( l , t ) results in a characteristic dependence of the @xmath28 cross - section on the @xmath29 polar angle of @xmath30 in the @xmath0 rest frame @xcite .
the interference terms between different helicities of the @xmath0 production are canceled if the cross section is integrated over the @xmath31 azimuthal decay angle of @xmath30 in the @xmath0 rest frame .
the total angular distribution @xmath32 , including the dependence on the @xmath30 polar angle , can be written separately for longitudinal @xmath33 and transverse @xmath34 mesons : @xmath35 .
\nonumber\end{aligned}\ ] ]
the data were accumulated with the hermes forward spectrometer during the running period 2002 - 2005 . the @xmath1 gev positron ( electron ) beam was scattered off a transversely polarized hydrogen target with an average polarization of @xmath36
. events with exactly one positron ( electron ) and two oppositely charged hadron tracks were selected .
exclusive @xmath0 events were identified by requiring @xmath37 gev , where @xmath38 is the missing mass squared and @xmath39 is the proton mass . due to the experimental resolution and limited acceptance ,
semi - inclusive pion production can contribute to the exclusive sample ; this is the primary background .
it is well reproduced by the pythia simulation and is estimated to be of the order of @xmath40 .
the ttsa asymmetry is extracted by using the unbinned maximum likelihood method where all the moments @xcite of @xmath41 , @xmath42 and @xmath42 ( eqs . [ eq : wut ] , [ eq : wut_sep ] ) are fitted simultaneously . in this analysis , the angular distributions @xmath43 and the asymmetries @xmath44 of @xmath0 , @xmath33 and @xmath34
meson productions are defined by unpolarized spin density matrix elements ( sdmes ) @xcite previously measured by hermes @xcite .
r0.6 the only ttsa moment of @xmath0s produced from longitudinal photons that is related to the gpds @xmath4 and @xmath5 , is the @xmath45 moment . in figure
[ fig : a_ut ] the @xmath46 moment of the ttsa is presented .
the panels show from left to right the integrated value and the @xmath15 , @xmath14 and @xmath20 dependences of the asymmetry . for the @xmath14 and @xmath20 dependences ,
@xmath15 is required to be above @xmath47 gev@xmath48 .
the upper panels represent the @xmath0 total asymmetries , while the middle and the lower panels represent the longitudinal @xmath33 and transverse @xmath34 separated asymmetries , respectively .
the error bars represent the statistical uncertainties only , while the yellow bands indicate the systematic uncertainties due to the target polarization , the background subtraction procedure , the uncertainty resulting from the the unpolarized sdmes measurement as well as the influence of the beam polarization on the final result .
the @xmath14 and @xmath20 dependences of the @xmath46 moment for longitudinal @xmath0 mesons are compared to the theoretical calculations @xcite ( see figure [ fig : a_ut_theor ] ) .
the longitudinal component of @xmath46 moment of the asymmetry is related to : @xmath49 , where the @xmath50 , @xmath51 and @xmath52 , @xmath53 represent the quark and gluon gpds , respectively .
currently no model exists for the gluon gpd @xmath52 . in the present theoretical calculations
the gluon gpd @xmath5 is neglected .
however , @xmath52 is not expected to be large compared to the quark gpds @xcite .
no large contribution is expected from sea quarks in our @xmath14 range . as gpd @xmath50
is related to the total angular momentum @xmath54 and @xmath55 carried by @xmath56 and @xmath57 quarks , the @xmath46 moment of the asymmetry is sensitive to @xmath54 and @xmath55 .
the various curves in figure [ fig : a_ut_theor ] represent those calculations for @xmath58 , @xmath59 and @xmath60 and @xmath61 .
the @xmath61 choice is motivated by the results of recent lattice calculation @xcite .
the comparison of @xmath14 and @xmath20 dependences of the asymmetry with theoretical calculations indicates that the data favors positive @xmath54 values . and
@xmath20 dependences of @xmath46 moment of the ttsa of exclusive production of @xmath33 mesons compared to the model calculations .
the error bars represent the total error.,width=207 ] and @xmath20 dependences of @xmath46 moment of the ttsa of exclusive production of @xmath33 mesons compared to the model calculations .
the error bars represent the total error.,width=207 ]
the @xmath46 moment of the ttsa of exclusive @xmath0 meson production is measured on a hydrogen target .
the kinematic dependences as well as the integrated value of the asymmetry are presented . in particular , the longitudinal part of the asymmetry is compared to theoretical calculations .
the model suggests that the data favors positive @xmath54 values , which is in agreement with deeply virtual compton scattering results obtained from hermes data @xcite .
99 d. mueller et al . , fortschr .
* 42 * ( 1994 ) 101 ; + a.v .
radyushkin , phys . rev .
* d55 * ( 1997 ) 7114 .
j. collins , l. frankfurt , m. strikman , phys .
* d56 * ( 1997 ) 2982 .
ji , phys .
* 78 * ( 1997 ) 610 .
k. goeke , m.v .
polyakov , m.vanderhaeghen , prog .
phus * 47 * , ( 2001 ) 401 .
a. bacchetta , u. dalesio , m. diehl , c.a .
miller , phys . rev .
* d70 * ( 2004 ) 117504 .
m. diehl , s. sapeta , eur.phys.j .
* c41 * ( 2005 ) 515 .
k. schilling , g. wolf , nucl .
b * 61 * ( 1973 ) 381 .
b. marianski , dis 2006 : proceedings of the 14th international workshop .
world scientific ( 2007 ) 255 .
f. ellinghaus , w .- d .
novak , a.v .
vinnikov , z. ye , eur.phys.j . *
c46 * ( 2006 ) 729 .
m. diehl , phys .
* 388 * ( 2003 ) 41 .
m.goekeler et al , phys .
* 92 * ( 2004 ) 042002 .
nowak , aip conf .
915 ( 2006 ) 603 . | preliminary measurements are reported on the azimuthal single - spin asymmetry of exclusive @xmath0 mesons for a transversely polarized hydrogen target at hermes using the @xmath1 gev hera positron beam . within the generalized parton distribution ( gpd )
formalism , this asymmetry is sensitive to the total angular momentum of quarks and gluons in the nucleon . since the gpd formalism is only valid for mesons produced by longitudinal photons , the transverse target - spin asymmetry of longitudinal @xmath0 mesons
is extracted assuming @xmath2-channel helicity conservation and compared to theoretical calculations . |
rovibrationally excited h@xmath0 molecules have been observed in many astrophysical objects ( for recent studies , see weintraub et al .
1998 ; van dishoeck et al . 1998 ; shupe et al .
1998 ; bujarrabal et al . 1998 ; stanke et al . 1998 ) .
the rovibrational levels of the molecule may be populated by ultraviolet pumping , by x - ray pumping , by the formation mechanism , and by collisional excitation in shock - heated gas ( dalgarno 1995 ) .
the excited level populations are then modified by collisions followed by quadrupole emissions .
the main colliding partners apart from h@xmath0 are h and he .
although he is only one tenth as abundant as h , collisions with he may have a significant influence in many astronomical environments depending on the density , temperature and the initial rotational and vibrational excitation of the molecule .
collisions with he and h@xmath0 are particularly important when most of the hydrogen is in molecular form , as in dense molecular clouds . to interpret observations of the radiation emitted by the gas ,
the collision cross sections and corresponding rate coefficients characterizing the collisions must be known .
emissions from excited rovibrational levels of the molecule provide important clues regarding the physical state of the gas , dissociation , excitation and formation properties of h@xmath0 . here
we investigate the collisional relaxation of vibrationally excited h@xmath0 by he .
rovibrational transitions in h@xmath0 induced by collisions with he atoms have been the subject of a large number of theoretical calculations in the past ( alexander 1976 , 1977 ; alexander and mcguire 1976 ; dove et al .
1980 ; eastes and secrest 1972 ; krauss and mies 1965 ; mcguire and kouri 1974 ; raczkowski et al .
1978 ) and continue to attract experimental ( audibert et al . 1976 ; michaut et al . 1998 ) and theoretical attention ( flower et al . 1998 ; dubernet & tuckey 1999 ; balakrishnan et al . 1999 ) .
recent theoretical calculations are motivated by the availability of more accurate representations of the interaction potentials and the possibility of performing quantum mechanical calculations with few approximations . the potential energy surface determined by muchnick and russek ( 1994 )
was used by flower et al .
( 1998 ) and by balakrishnan et al .
( 1999 ) in recent quantum mechanical calculations of rovibrational transition rate coefficients for temperatures ranging from 100 to 5000k .
flower et al . presented their results for vibrational levels @xmath3 and 2 of ortho- and para - h@xmath0 .
balakrishnan et al . ( 1999 ) reported similar results for @xmath4 and 1 . though both authors have adopted similar close - coupling approaches for the scattering calculations , flower et al .
used a harmonic oscillator approximation for h@xmath0 vibrational wave functions in evaluating the matrix elements of the potential while the calculations of balakrishnan et al .
made use of the h@xmath0 potential of schwenke ( 1988 ) and the corresponding numerically determined wave functions .
the results of the two calculations agreed well for pure rotational transitions but some discrepancies were seen for rovibrational transitions .
we believe this may be due to the different choice of vibrational wave functions .
the sensitivity of the rate coefficients to the choice of the h@xmath0 wave function was noted previously and differences could be significant for excited vibrational levels .
we find this to be the case for transitions involving @xmath5 .
thus , in this article , we report rate coefficients for transitions from @xmath6 to 6 initial states of h@xmath0 induced by collisions with he atoms using numerically exact quantum mechanical calculations .
we also report results of quasiclassical trajectory ( qct ) calculations and examine the suitability of classical mechanical calculations in predicting rovibrational transitions in h@xmath0 .
the quantum mechanical calculations were performed using the nonreactive scattering program molscat developed by hutson and green ( 1994 ) with the he - h@xmath0 interaction potential of muchnick and russek ( 1994 ) and the h@xmath0 potential of schwenke ( 1988 ) .
we refer to our earlier paper ( balakrishnan , forrey & dalgarno , 1999 ) for details of the numerical implementation .
different basis sets were used in the calculations for transitions from different initial vibrational levels .
we use the notation [ @xmath7@xmath8(@xmath9@xmath10 ) to represent the basis set where the quantities within the square brackets give the range of vibrational levels and those in braces give the range of rotational levels coupled in each of the vibrational levels . for transitions from @xmath11 and 4 we used , respectively , the basis sets [ 03](011 ) & [ 4](03 ) , [ 03](011 ) & [ 4](09 ) and [ 35](011 ) & [ 1,6](011 ) . for @xmath12 and 6 of para h@xmath0 we used , respectively , [ 46](014 ) & [ 3,7](08 ) and [ 57](014 ) & [ 4,8](08 ) . during the calculations
, we found that the @xmath13 transitions are weak with cross sections that are typically orders of magnitude smaller than for the @xmath14 transitions .
thus , for @xmath12 and 6 of ortho - h@xmath0 , we have only included the @xmath14 vibrational levels with @xmath15=013 in the basis set to reduce the computational effort .
the basis sets were chosen as a compromise between numerical efficiency and accuracy and could introduce some truncation errors for transitions to levels which lie at the outer edge of the basis set .
our convergence tests show that truncation errors are small .
rovibrational transition cross sections @xmath16 where the pairs of numbers @xmath17 and @xmath18 respectively denote the initial and final rovibrational quantum numbers , were computed for kinetic energies ranging from 10@xmath1 to 3 ev .
sufficient total angular momentum partial waves were included in the calculations to secure convergence of the cross sections .
the quasiclassical calculations were carried out using the standard classical trajectory method as described by lepp , buch and dalgarno ( 1995 ) in which the procedure of blais and truhlar ( 1976 ) was adopted for the final state analysis .
because rovibrational transitions are rare at low velocities , useful results could be obtained only for collisions at energies above 0.1 ev .
the results are averages over 10000 trajectories .
the quantum mechanical calculations were performed using molscat ( hutson & green , 1994 ) suitably adapted for the present system with the potential represented by a legendre polynomial expansion in which we retained nonvanishing terms of orders 0 to 10 , inclusive . calculation of rovibrational transition rate coefficients over a wide range of temperatures requires the determination of scattering cross sections at the energies spanned by the boltzmann distribution at each temperature .
this is a computationally demanding problem especially when quantum mechanical calculations are required . for many systems ,
qct calculations offer a good compromise between accuracy and computational effort .
however , the validity of classical mechanics is in question especially for lighter systems and at lower temperatures where quantum mechanical effects such as tunneling are important . due to the small masses of the atoms involved ,
the present system offers an excellent opportunity to test the reliability of qct calculations in predicting rovibrational transitions .
though such attempts have been made in the past ( dove et al .
1980 ) the classical mechanical and quantum mechanical calculations were done in different energy regimes and a one - to - one comparison was not possible .
we carried out quantum mechanical and qct calculations of rovibrational transition cross sections for the present system over a wide range of energies . in figure 1
we compare the results for pure rotational de - excitation transitions ( @xmath19 ) with @xmath20 in @xmath4 to 5 .
there are striking similarities and differences between the quantum mechanical and qct results .
they agree quite well at higher energies and have a similar energy dependence with both calculations predicting the same maximum for the cross sections before they fall off .
in contrast , the agreement becomes less satisfactory at lower energies with the qct cross sections rapidly decreasing to zero as the energy is decreased .
the quantum mechanical results pass through minima and subsequently increase with further decrease of the kinetic energy .
this is a purely quantum mechanical effect that has important consequences for the low temperature rate coefficients ( balakrishnan et al .
the quantum mechanical cross sections eventually conform to an inverse velocity dependence as the relative translational energy is decreased to zero and the corresponding rate coefficients are finite in the limit of zero temperature , in accordance with wigner s threshold law .
the results illustrate that qct calculations may be used reliably to calculate rotational transitions for energies higher than 0.5 ev , but at lower energies , quantum mechanical calculation must be employed .
the energy regime for the validity of qct results is more restricted for rovibrational transitions as illustrated in figure 2 in which we show cross sections for rovibrational transitions @xmath21 with @xmath22 in @xmath23 to 5 .
the results indicate that the qct method is inadequate to calculate rovibrational transition cross sections at impact energies below 1 ev .
the qct results exhibit a sharp fall below a threshold near an energy of 1 ev whereas the quantum mechanical results vary smoothly .
similar results hold for other transitions .
the higher collision energies required for the validity of the qct method for rovibrational transitions compared to pure rotational transitions is due in part to the much smaller cross sections for rovibrational transitions making the results more sensitive to the details of the dynamics .
rate coefficients @xmath24 were calculated for the temperature range @xmath25 to 4000 k by averaging the cross sections over a boltzmann distribution of relative velocities .
flower et al .
( 1998 ) have reported rate coefficients for rovibrational transitions from @xmath26 for @xmath27 and 4500 k. a comparison of some of the de - excitation rate coefficients from the @xmath6 level calculated in this paper and those reported by flower _
is given in table 1 for @xmath28 k. the agreement is good for rovibrational transitions involving @xmath29 and 4 but larger differences , by factors between 2 and 4 , are seen for transitions involving larger values of @xmath30 .
our rate coefficients are generally greater than those of flower _ et al . _
( 1998 ) for transitions where the discrepancy is large .
our rate coefficient calculations extend those of flower _ et al . _
( 1998 ) to include the @xmath31 and 6 vibrational levels .
the preferential formation of h@xmath0 in these vibrational levels has been discussed by dalgarno ( 1995 ) . in tables 2 and 3 , we present our comprehensive results for rovibrational de - excitation transitions from different rotational levels in para- and ortho - h@xmath0 from the @xmath6 level as functions of temperature .
the corresponding excitation rate coefficients may be obtained from detailed balance @xmath32 } k_{vj , v'j'}(t)\ ] ] where @xmath33 is the boltzmann constant and @xmath34 is the rovibrational energy of the molecule in the @xmath35 level .
similar results for @xmath23 to 6 are presented in tables 4 - 11 .
tables 2 - 11 reveal some interesting aspects of energy transfer .
it can be seen that for temperatures less than 1000 k , rate coefficients for rovibrational transitions involving @xmath36 predominate over other transitions where changes in both @xmath37 and @xmath15 occur .
this is clearly seen for ( @xmath38 ) transitions with @xmath39 for @xmath40 and the effect becomes stronger with increasing rotational excitation but less important as @xmath41 increases .
this is an example of quasi - resonant scattering ( stewart et al .
1988 ; forrey et al .
a detailed study of this process in the limit of zero temperature and its correspondence with classical mechanics has been carried out recently ( forrey et al . 1999 ) .
the efficient conversion of vibrational energy into rotational energy may produce a significant population of high rotational levels in environments where molecular hydrogen is subjected to an intense flux of x - rays or ultraviolet photons .
this work was supported by the national science foundation ( nsf ) , division of astronomy .
mv was supported by the nsf through the research experience for undergraduates program at the smithsonian astrophysical observatory .
hutson , j.m .
, & green , s. 1994 , molscat version 14 distributed by collaborative computational project 6 ( daresbury laboratory : uk engineering and physical sciences research council ) .
krauss , m. , & mies , f.h .
1965 , j. chem .
phys . , 42 , 2703 .
lepp , s. , buch , v. , & dalgarno , a 1995 , apjs , 98 , 345 .
mcguire , p. , & kouri . , d. j. , 1974 , j. chem .
, 60 , 2488 .
michaut , x. , saint - loup , r. , berger , h. , dubernet , m. l. , joubert , p. , & bonamy , j. 1998 , j. chem .
109 , 951 .
muchnick , p. , & russek , a. 1994 , j. chem .
phys . , 100 , 4336 .
raczkowski , a.w . ,
lester , w.a .
jr . , & miller , w.h . 1978 , j. chem .
phys . , 69 , 2692 .
schwenke , d. w. , 1988 , j. chem .
phys . , 89 , 2076 . | we present quantum mechanical and quasiclassical trajectory calculations of cross sections for rovibrational transitions in ortho- and para - h@xmath0 induced by collisions with he atoms .
cross sections were obtained for kinetic energies between 10@xmath1 and 3 ev , and the corresponding rate coefficients were calculated for the temperature range 100@xmath24000 k. comparisons are made with previous calculations . |
quantum information science ( see e.g. @xcite and references therein ) has received an increased attention in recent years due to the understanding that it enables to perform procedures not possible by purely classical resources .
experimental techniques to manipulate increasingly complex quantum systems are also rapidly progressing .
one of the central issues is on the one hand to control and manipulate delicate complex quantum states in an efficient manner , but on the other hand at the same time to prevent all uncontrollable influences from the environment . in order to tackle such problems ,
one has to understand the structure and properties of quantum states .
this can be done either through studies of particular states in a particular setting , or through focusing on the properties of the most generic states .
random quantum states , that is states distributed according to the unitarily invariant fubini - study measure , are good candidates for describing generic states .
indeed , they are typical in the sense that statistical properties of states from a given hilbert space are well described by those of random quantum states . also , they describe eigenstates of sufficiently complex quantum systems @xcite as well as time evolved states after sufficiently long evolution .
not least , because random quantum states possess a large amount of entanglement they are useful in certain quantum information processes like quantum dense coding and remote state preparation @xcite . they can be used to produce random unitaries needed in noise estimation @xcite and twirling operations @xcite .
in addition , as random states are closely connected to the unitarily invariant haar measure of unitary matrices , the unitary invariance makes theoretical treatment of such states simpler .
producing such states therefore enables to make available a useful quantum resource , and in addition to span the space of quantum states in a well - defined sense .
therefore several works have recently explored different procedures to achieve this goal .
it is known that generating random states distributed according to the exact invariant measure requires a number of gates exponential in the number of qubits .
a more efficient but approximate way to generate random states uses pseudo - random quantum circuits in which gates are randomly drawn from a universal set of gates .
as the number of applied gates increases the resulting measure gets increasingly close to the asymptotic invariant measure @xcite .
some bipartite properties of random states can be reproduced in a number of steps that is smaller than exponential in the number of qubits .
polynomial convergence bounds have been derived analytically for bipartite entanglement @xcite for a number of pseudo - random protocols . on the numerical side , different properties of circuits generating random states
have been studied @xcite . in order to quantify
how well a given pseudo - random scheme reproduces the unitarily invariant distribution , one can study averages of low - order polynomials in matrix elements @xcite .
in particular , one can define a state or a unitary @xmath0-design , for which moments up to order @xmath0 agree with the haar distribution @xcite .
although exact state @xmath0-designs can be built for all @xmath0 ( see references in @xcite ) they are in general inefficient . in contrast , efficient approximate @xmath0-designs can be constructed for arbitrary @xmath0 ( for the specific case of 2-design see @xcite ) .
the pseudo - random circuit approach can yield only pseudo - random states , which do not reproduce exactly the unitarily invariant distribution .
the method has been shown to be useful for large number of qubits , where exact methods are clearly inefficient .
however , for systems with few qubits , the question of asymptotic complexity is not relevant .
it is thus of interest to study specifically these systems and to find the most efficient way in terms of number of gates to generate random states distributed according to the unitarily invariant measure .
this question is not just of academic interest since , as mentioned , few - qubit random unitaries are needed for e.g. noise estimation or twirling operations .
optimal circuits for small number of qubits could also be used as a basic building block of pseudo - random circuits for larger number qubits , which might lead to faster convergence .
in addition , systems of few qubits are becoming available experimentally , and it is important to propose algorithms that could be implemented on such small quantum processors , and which use as little quantum gates as possible .
indeed , quantum gates , and especially two - qubit gates , are a scarce resource in real systems which should be carefully optimized . in this paper
we therefore follow a different strategy from the more generally adopted approach of using pseudo - random circuits to generate pseudo - random states , and try and construct exact algorithms generating random states for systems of three qubits . in the language of @xmath0-designs
such algorithms are exact @xmath1-designs .
we present a circuit composed of one - qubit and two - qubit gates which produces exact random states in an optimal way , in the sense of using the smallest possible number of cnot gates .
the circuit uses in total @xmath2 cnot gates and @xmath3 one - qubit elementary rotations .
our procedure uses results recently obtained @xcite which described optimal procedures to transform a three - qubit state into another .
our circuit needs @xmath4 random numbers which should be classically drawn and used as parameters for performing the one - qubit gates .
the probability distribution of these parameters is derived , showing that it factorizes into a product of 10 independent distributions of one parameter and a joint distribution of the 4 remaining ones , each of these distributions being explicitly given .
since we had to devise specific methods to compute these distributions , we explain the derivation in some details , as these methods can be useful in other contexts . after presenting the main idea of the calculation in section [ metrictensor ] , we start by treating the simple case of two - qubit states in section [ 2qubits ] .
we then turn to the three - qubit case and first show factorization of the probability distribution for a certain subset of the parameters ( section [ 7 - 14 ] ) , the remaining parameters being treated in section [ 1 - 6 ] .
the full probability distribution for three qubits is summarized in section [ explicit_calc ] .
formally , a quantum state @xmath5 can be considered as an element of the complex projective space @xmath6 , with @xmath7 the hilbert space dimension for @xmath8 qubits @xcite .
the natural riemannian metric on @xmath6 is the fubini - study metric , induced by the unitarily invariant haar measure on @xmath9 .
it is the only metric invariant under unitary transformations . to parametrize @xmath6
one needs @xmath10 independent real parameters . such parametrizations are well - known , for instance using hurwitz parametrization of @xmath9 @xcite .
however , they do not easily translate into one and two - qubit operations , as desired in quantum information . in ref .
@xcite , optimal quantum circuits transforming the three - qubit state @xmath11 into an arbitrary quantum state were discussed . in the case of three qubits
, a generic state can be parametrized up to a global phase by 14 parameters .
the quantum circuit requiring the smallest amount of cnot gates has three cnots and 15 one - qubit gates depending on 14 independent rotation angles
. from @xcite it is possible ( see appendix ) to extract the circuit depicted in fig .
[ circuit ] , expressed as a series of cnot gates and single qubit rotations , where z - rotation is @xmath12 and y - rotation is @xmath13 with @xmath14 the pauli matrices .
the circuit allows to go from @xmath11 to any quantum state ( up to an irrelevant global phase ) .
it therefore provides a parametrization of a quantum state @xmath5 by angles @xmath15 . in order to generate random vectors distributed according to the fubini - study measure
, it would of course be possible to use e. g. hurwitz parametrization to generate classically a random state , and then use the procedure described in @xcite to find out the consecutive steps that allow to construct this particular vector from @xmath11 .
however this procedure requires application of a specific algorithm for each realization of the random vector . instead , our aim here is to directly find the distribution of the @xmath16 such that the resulting @xmath5 is distributed according to the fubini - study measure .
this is equivalent to calculating the invariant measure associated with the parametrization provided by fig .
[ circuit ] in terms of the angles @xmath15 .
geometrically , the fubini - study distance @xmath17 is the angle between two normalized states , @xmath18 .
the metric induced by this distance is obtained by taking @xmath19 , getting @xmath20 where @xmath21 is the usual hermitian scalar product on @xmath22 @xcite .
if a state @xmath5 is parametrized by some parameters @xmath23 then the riemannian metric tensor @xmath24 is such that @xmath25 and the volume form at each point of the coordinate patch , directly giving the invariant measure , is then given by @xmath26 .
thus the joint distribution @xmath27 of the @xmath16 is simply obtained by calculating the determinant of the metric tensor given by with the parametrization @xmath28 , @xmath29 .
unfortunately the calculation of such a @xmath30 determinant for @xmath31 qubits is intractable and one has to resort to other means .
let us first consider the easier case of @xmath32 qubits , where by contrast the calculation can be performed directly .
a normalized random 2-qubit state @xmath5 depends , up to a global phase , on 6 independent real parameters .
a circuit producing @xmath5 from an initial state @xmath33 is depicted in fig .
[ fig:2qcircuit ] .
one can easily calculate the parametrization of the final state @xmath34 in terms of all six angles , thus directly obtaining the metric tensor @xmath24 from ( [ ds2 ] ) .
square root of the determinant of @xmath35 then gives an unnormalized probability distribution of the angles as @xmath36 ( see also @xcite ) .
several observations can be made about this distribution .
first , the three rotations applied on the first qubit ( top wire in fig . [ fig:2qcircuit ] ) after the cnot gate represent a random su(2 ) rotation , for which the y - rotation angle is distributed as @xmath37 and the z - rotation angles are uniformly distributed @xcite .
secondly , angle @xmath38 gives the eigenvalue of the reduced density matrix , @xmath39 , for which the distribution is well known , see , _
e.g. _ , @xcite .
the third observation is that , remarkably , the joint distribution ( [ eq:2qp ] ) of all 6 angles factorizes into 6 independent one - angle distributions .
let us now turn to our main issue , which is the distribution of angles in the three - qubit case . in order to have an indication whether the distribution of an angle @xmath16 factorizes , we numerically computed the determinant det@xmath42 of the metric tensor as a function of @xmath16 with the other angles fixed .
we also numerically computed the marginal distribution of @xmath16 by using the procedure given in the appendix to find the angles corresponding to a sample of uniformly distributed random vectors . if the distribution for a given angle @xmath16 factorizes , these two numerically computed functions should match ( up to a constant factor ) .
this is what we observed for all angles but four of them ( angles @xmath43 to @xmath44 ) . in order to turn this numerical observation into a rigorous proof
, we are going to show in this section that the distributions for angles @xmath45 to @xmath41 indeed factorize . in the next section
we will complete the proof by dealing with the cases @xmath38 to @xmath44 .
the explicit analytical expression of the probability distribution for individual angles will be given in section [ explicit_calc ] .
let us denote by @xmath46 the circuit of fig .
[ circuit ] and by @xmath47 the unitary operator corresponding to it , so that @xmath48 . because circuits @xmath46 span the whole space of 3-qubit states , any unitary 3-qubit transformation @xmath49 maps parameters @xmath50 to new parameters @xmath51 such that @xmath52 .
we denote by @xmath53 the circuit parametrized by angles @xmath51 corresponding to performing @xmath54 followed by @xmath49 .
it is associated with the unitary operator @xmath55 such that @xmath56 .
unitary invariance of the measure implies for @xmath27 that @xmath57 with @xmath58 the jacobian of the transformation @xmath59 and @xmath60 denotes the determinant .
note that eq .
( [ eq : invariance ] ) is not a simple change of variables , as the same function @xmath61 appears on both sides of the equation .
the jacobian matrix @xmath58 for transformation @xmath49 from angles @xmath50 to @xmath51 , @xmath52 , tells how much do the angles @xmath51 of @xmath62 change if we vary angles @xmath50 in @xmath63 keeping transformation matrix @xmath49 fixed .
choosing @xmath49 that sets some angles @xmath64 in circuit @xmath53 to a fixed value , say zero , and at the same time showing that @xmath65 depends only on these angles @xmath64 , would prove factorization of @xmath27 with respect to angles @xmath64 through eq .
[ eq : invariance ] .
the simplest case is that of gates at the end of the circuit @xmath54 of fig .
[ circuit ] , _
e.g. _ , gate @xmath66 .
for @xmath49 we take y - rotation by angle @xmath67 on the third qubit , @xmath68 .
it defines a mapping @xmath59 such that @xmath69 for @xmath70 and @xmath71 .
matrix elements of the jacobian , i.e. partial derivatives @xmath72 , are equal to the kronecker symbol @xmath73 .
the jacobian is equal to an identity matrix and its determinant is one .
equation taken at @xmath74 then gives @xmath75 , from which one concludes that the distribution for @xmath41 factorizes and is in fact uniform ( unless noted otherwise @xmath61 s are not normalized ) .
the same argument holds for the two other rotations by angles @xmath76 and @xmath77 applied at the end of each qubit wire .
proceeding to angle @xmath78 one could use @xmath79 applied on the first qubit and show that the jacobian depends only on @xmath78 and @xmath80 , while at @xmath81 and @xmath82 one gets @xmath83 , from which factorization of @xmath78 would follow from eq .
[ eq : invariance ] .
there is however a simpler way .
observe that the three single - qubit gates with angles @xmath84 and @xmath80 on the first qubit span the whole su(2 ) group .
therefore , for any one - qubit unitary @xmath49 , gates @xmath85 can be rewritten as @xmath86 , without affecting other @xmath87 s .
the distribution of these three angles must therefore be the same as the distribution of corresponding su(2 ) parameters .
note that the same argument can be applied in the 2-qubit case of section [ 2qubits ] . as a consequence
, the distribution of angles for gates @xmath88 at the end of the circuit should be the same in both cases , that is the distribution of @xmath40 is uniform while that of @xmath78 is proportional to @xmath89 .
similarly , one can show that the distribution for the angles @xmath90 to @xmath76 is the same as for angles @xmath40 to @xmath80 .
as opposed to gates 10 - 12 , for gate 13 we can not use the analogy with the 2-qubit circuit ( fig.[fig:2qcircuit ] ) because the two gates 13 and 14 on the third qubit do not span the whole su(2 ) group . therefore a different argument should be used . in what follows
we show that the joint distribution for angles @xmath91 and @xmath41 can be factorized out of the full distribution .
since it has been shown in subsection [ gate14 ] that angle @xmath41 factorizes , this will prove that the distribution for @xmath91 also factorizes .
using @xmath92 on the third qubit we can set @xmath93 and @xmath94 to zero with the choice @xmath95 and @xmath96 .
our goal is to show that @xmath65 depends only on @xmath91 and @xmath41 .
we can formally consider each angle @xmath97 as being a function @xmath98 of the initial @xmath50 as well as of the parameters @xmath99 through @xmath100 . to calculate matrix elements of @xmath58 for our choice of @xmath49
evaluated at @xmath95 and @xmath96 , we must obtain the first - order expansion in @xmath101 of the quantities @xmath102 some angles @xmath103 are very simple .
we immediately see that when varying angles @xmath104 , that is taking @xmath105 in eq.[limlim ] , angles @xmath106 do not change ( i.e @xmath107 ) .
the corresponding @xmath108-dimensional subblock in @xmath58 is therefore equal to an identity matrix .
similarly , taking @xmath109 in eq.[limlim ] we see that @xmath110 do not change .
the corresponding column in @xmath58 is therefore zero apart from @xmath111 on the diagonal .
the jacobian thus has a block structure of the form @xmath112 where @xmath113 is a @xmath114-dimensional identity matrix and @xmath115 is a @xmath116-dimensional block with partial derivative @xmath117 .
the angle @xmath94 given by eq.([limlim ] ) is obtained by varying angle @xmath41 by @xmath101 . the condition that @xmath100 is @xmath118 , where @xmath119 is a state after the third cnot acts on @xmath120 ( counting from the left in fig . [ circuit ] ) , explicitly given by @xmath121 with @xmath122 @xmath123 is therefore determined by @xmath124 and @xmath125 , where @xmath126 .
similarly , @xmath127 is determined by @xmath128 and @xmath129 .
projecting these conditions for @xmath130 and @xmath131 on the computational basis @xmath132 , eliminating unwanted variables , we get the following two equations : @xmath133 one with @xmath134 , one with @xmath135 . expanding these equations to first order in @xmath101 yields @xmath136 and @xmath137 .
the derivative @xmath138 is therefore equal to @xmath139 , which completes the proof .
incidentally , we also see that the distribution of @xmath91 is proportional to @xmath140 .
in the preceding section , we have shown that the distribution for angles @xmath40 to @xmath41 factorizes . as was mentioned , numerical observations indicated us that the distribution for angles @xmath38 and @xmath141 should also factorize , but that it is not the case for the joint distribution of @xmath142 .
as we were not able to directly prove by the same methods as above that the distributions for @xmath38 and @xmath141 factorize , we use a different strategy .
namely , we first assume that this factorization is true , then we compute the distributions under this assumption , and the knowledge of the answer allows us to prove a posteriori that it is indeed the correct probability distribution . if the factorization holds , the distribution for @xmath38 and @xmath141 is easily calculated from the matrix @xmath35 using symbolic manipulation software , by replacing angles @xmath64 , @xmath143 , in @xmath35 by suitably chosen simple values , so that the @xmath30 determinant giving the volume form can now be handled .
this yields , up to a normalization constant , @xmath144 the joint distribution of @xmath142 can not be further factorized , and requires heavy calculations . indeed , even replacing all angles but @xmath145 by numerical values the determinant @xmath146 of the metric tensor given by still depends on 4 variables , which is too much for it to be evaluated by standard software .
we thus proceed as follows .
first one can show that @xmath146 can be put under the form @xmath147 with the sums running over all @xmath148 but only even values of @xmath149 and @xmath150 .
because of the parity of @xmath151 , there are @xmath152 independent coefficients @xmath153 .
evaluating numerically the determinant at @xmath154 random values of the angles one gets an @xmath155 linear system that can be solved numerically .
if the values of the coefficients of the matrix @xmath24 are multiplied by a factor 4 , then one is ensured ( from inspection of @xmath146 ) that the @xmath153 are rationals of the form @xmath156 , @xmath157 .
this allows to deduce their exact value from the numerical result .
we are left with 6998 nonzero terms in @xmath146 , and terms with odd @xmath158 or @xmath159 do not exist .
we then suppose that @xmath160 can be expanded as @xmath161 this assumption is validated a posteriori , since a solution of the form can indeed be found .
there are 4851 coefficients @xmath162 , which can be obtained by identifying term by term coefficients in the expansion of @xmath163 and @xmath146 .
we have to solve a system of quadratic equations @xmath164 the first equation is quadratic and fixes an overall sign .
equation @xmath165 is linear once the values obtained from equations @xmath111 to @xmath0 are plugged into it .
starting with the highest - degree term @xmath166 one can thus recursively solve all equations .
there are only 1320 non - zero coefficients @xmath162 .
gathering together terms @xmath167 one can simplify the sum to a sum of 96 terms of the form @xmath168 . expanding this expression in powers of @xmath169 and @xmath170 and simplifying separately each coefficient we finally get @xmath171 where @xmath172 and @xmath173 with @xmath174 , @xmath130 given by eq . .
recall that @xmath131 is the bit - flip transform of @xmath130 , @xmath126 .
note that @xmath175 .
angles @xmath176 can be obtained from @xmath177 where @xmath178 and @xmath179 .
we do not have a general argument to explain this remarkable expression of the distribution in terms of the scalar products of @xmath174 , @xmath130 and @xmath131 . to complete the proof for the joint distribution @xmath180 it remains to be checked that the determinant of the metric tensor @xmath35 with angles @xmath40 to @xmath41 replaced by constants is indeed proportional to @xmath181 .
this a posteriori verification is easier to handle symbolically than the full a priori calculation of the @xmath30 determinant .
indeed , the determinant can first be reduced to an @xmath182 determinant by gauss - jordan elimination .
the remaining determinant can be expanded as a trigonometric polynomial . although symbolic manipulation software do not allow to simplify the coefficients of this polynomial , they are able to check that these coefficients match those of the expected distribution .
we proved in that way that the difference between the determinant det@xmath42 and our expression is identically zero .
this gives a computer - assisted but rigorous proof for the distribution of angles @xmath38 to @xmath44 .
gathering together the results of the previous sections we obtain that the joint distribution @xmath27 can be factorized as @xmath183 the joint distribution @xmath184 has been derived in the previous section and is given by eq . .
the distribution for @xmath38 and @xmath141 is given by eqs . - .
given the factorization , it is easy to calculate the remaining @xmath185 for each @xmath186 as was done for @xmath38 and @xmath141 in the previous section : replacing angles @xmath64 , @xmath187 , in @xmath35 by suitably chosen simple values , the @xmath30 determinant giving the volume form can be easily evaluated by standard symbolic manipulation .
this yields , up to a normalization constant , @xmath188 the knowledge of the angle distribution ( [ eq : p ] ) allows to easily generate random three - qubit vectors using the circuit of fig .
[ circuit ] .
angles @xmath189 and @xmath40 to @xmath41 can be drawn classically according to their individual probability distribution .
angles @xmath190 can be obtained classically from the joint distribution by , for instance , monte - carlo rejection method ( that is , drawing angles @xmath43 to @xmath44 and a parameter @xmath191 $ ] at random , and keeping them if @xmath192 ) . bounding @xmath184 from above by @xmath193 yields a success rate of about @xmath194 .
in this work , we constructed a quantum circuit for generating three - qubit states distributed according to the unitarily invariant measure .
the construction is exact and optimal in the sense of having the smallest possible number of cnot gates .
the procedure requires a set of @xmath4 random numbers classically drawn , which will be the angles of the one - qubit rotations , and whose distribution has been explicitly given .
remarkably , we have shown that the distribution of angles factorizes , apart from that of four angles .
the circuit can be used as a three - qubit random state generator , thus producing at will typical states on three qubits .
it could be also used as a building block for pseudo - random circuits in order to produce pseudo - random quantum states on an arbitrary number of qubits . at last
, it gives an example of a quantum algorithm producing interesting results which could be implemented on a few - qubit platform , using only @xmath195 quantum gates , of which @xmath3 are one - qubit elementary rotations much less demanding experimentally .
we thank the french anr ( project infosysqq ) and the ist - fet program of the ec(project eurosqip ) for funding .
m would like to acknowledge support by slovenian research agency , grant j1 - 7437 , and hospitality of laboratoire de physique th ' eorique , toulouse , where this work has been started .
in this appendix , we explain how to obtain the angles @xmath16 of the circuit ( fig . [ circuit ] ) for a given @xmath5 , based on the discussion in @xcite .
this justifies the use of these angles as a parametrization of the quantum states .
we start from a state @xmath5 , and transform it by the inverse of the different gates of fig . [ circuit ] to end up with @xmath11 , specifying how the angles @xmath16 are obtained in turn .
more details can be found in @xcite .
a generic three - qubit state @xmath5 can be written in a canonical form as a sum of two ( not normalized ) product terms @xcite , @xmath196 where @xmath197 are one - qubit states , @xmath198 is a one - qubit state orthogonal to @xmath199 and @xmath200 is a two - qubit state of the second and third qubits . the angle @xmath80 is chosen such that the z - rotation of angle @xmath201 eliminates a relative phase between the coefficients of the expansion of @xmath199 into @xmath202 and @xmath203 .
( note that because we are using the circuit in the reverse direction the angles of rotations have opposite signs ) . a subsequent y - rotation with angle @xmath204 results in the transformation @xmath205 ( up to a global phase ) .
similarly , rotations of angles @xmath206 and @xmath207 rotate @xmath208 into @xmath202 . after applying rotations of angles @xmath204 , @xmath201 , @xmath207 and @xmath206 the state has become of the form @xmath209 ( up to normalization ) .
two rotations on the third qubit of angles @xmath210 and @xmath211 are now chosen so as to rotate @xmath212 into some new state @xmath213 while @xmath214 is rotated , up to normalization , into @xmath215 .
it was shown in @xcite that this can always be done by writing the normalized @xmath216 as @xmath217 , and then @xmath41 is a solution of @xmath218 while @xmath219 , where @xmath220 s are relative phases in @xmath221 .
acting with a cnot@xmath222 gate on the resulting state one obtains a quantum state for the three qubits of the form @xmath223 , with @xmath224 .
the z - rotation angle @xmath225 on the second qubit is now determined so as to eliminate a relative phase between the expansion coefficients of @xmath226 , making them real up to a global phase .
on the third qubit we now apply three rotations of angles @xmath227 , @xmath228 , and @xmath229 to bring @xmath230 to @xmath231 and @xmath232 into @xmath233 , eliminating also a relative phase . then a cnot@xmath234 is applied . at this point
( after the second cnot in fig . [ circuit ] , counting from right , but without the @xmath40 rotation )
, the state has become of the form @xmath235 , where the one - qubit states @xmath236 and @xmath237 are normalized and real . with @xmath40
we now eliminate the relative phase @xmath238 , and with an y - rotation of angle @xmath239 the third qubit is brought to the state @xmath202
. then the combination of two y - rotation of angles @xmath141 and @xmath240 with a cnot@xmath241 brings the second qubit to @xmath202 , and the last rotation of angle @xmath242 on the first qubit yields the final state @xmath11 .
note that in the circuit of fig .
[ circuit ] the two z - rotations of angles @xmath40 and @xmath90 commute with cnot gates if they act on the control qubit .
this is the reason why the rotation of angle @xmath40 can be applied at any point between @xmath38 and @xmath78 and , similarly , @xmath90 can be applied at any point between @xmath141 and @xmath243 .
weinstein and c. s. hellberg , phys .
. lett . * 95 * , 030501 ( 2005 ) ; m. nidari , phys .
a * 76 * , 012318 ( 2007 ) ; y. most , y. shimoni and o. biham , phys .
a * 76 * , 022328 ( 2007 ) ; d. rossini and g. benenti , phys .
100 * , 060501 ( 2008 ) ; y. s. weinstein , w. g. brown and l. viola , phys .
a * 78 * , 052332 ( 2008 ) ; l. arnaud and d. braun , phys .
a * 78 * , 062329 ( 2008 ) . | we explicitly construct a quantum circuit which exactly generates random three - qubit states .
the optimal circuit consists of three cnot gates and fifteen single qubit elementary rotations , parametrized by fourteen independent angles .
the explicit distribution of these angles is derived , showing that the joint distribution is a product of independent distributions of individual angles apart from four angles . |
the force between uncharged conducting surfaces , the so - called `` casimir force , '' has been described as one of the least intuitive consequences of quantum electrodynamics . for conducting parallel flat plates separated by distance @xmath0 ,
this force per unit area @xmath1 has the magnitude @xcite : @xmath2 this relationship can be derived by considering the electromagnetic mode structure between the two plates , as compared to free space , and by assigning a zero - point energy of @xmath3 to each electromagnetic mode ( photon ) .
the change in total energy density between the plates , as compared to free space , as a function of separation @xmath4 , leads to the force of attraction .
the only fundamental constants that enter eq .
( 1 ) are @xmath5 and @xmath6 ; the electron charge @xmath7 is absent , implying that the electromagnetic field is not coupling to matter .
the role of @xmath6 is to convert the electromagnetic mode wavelength to a frequency , while @xmath5 converts the frequency to an energy .
the term `` casimir effect '' is applied to a number of long - range interactions , such as those between atoms or molecules ( retarded van der waals interaction ) , an atom and a material surface ( casimir - polder interaction ) , and the attraction between bulk material bodies . the latter effect is generally referred to as `` the '' casimir force and depends only on the bulk properties of the bodies under consideration ; i will limit my discussion to the latter effect . for real materials , eq .
( 1 ) must break down when the separation @xmath4 is so small that the mode frequencies are higher than the plasma frequency ( for a metal ) or higher than the absorption resonances ( for a dielectric ) of the material used to make the plates ; for sufficiently small separation , the force of attraction varies as @xmath8 , as discussed in particular by lifshitz @xcite . in analogy with the attractive forces between atoms ,
the force in this range is sometimes referred to as the london - van der waals attraction , while the @xmath9 range is referred to as the retarded van der waals ( casimir ) interaction . for the casimir effect , the crossover distance between the regimes is @xmath10 nm , much larger than atomic spacings in the materials , so it still makes sense to describe the materials by their bulk properties ( index of refraction ) ; the @xmath8 vs. @xmath9 interaction is in this case due to the truncation of the mode frequencies that are affected by the changing plate separation .
therefore the crossover between the two regimes appears to be of physically different origin than in the case of the attractive forces between isolated atoms .
the casimir effect and its calculation represent an electromagnetic waveguide problem , where imperfect materials are used in the construction of the waveguide .
the zero - point fields are those associated with the waveguide ; these modes do not exist in free space , so the idea that the casimir effect represents a negative energy density compared to free space is incorrect . given that the distances where the force of attraction is sufficiently strong to be experimentally detected are @xmath11 nm or less , an accurate theoretical description of an experimental system must take into account the real material properties as will be discussed in sec .
iii below . finally , it is interesting to note that there have been only a dozen or so published experimental measurements of the casimir force , to be compared with the hundreds and hundreds of theoretical papers on the subject .
perhaps very few doubt the strict validity of eq .
( 1 ) or its modification for real materials @xcite . because of the unavoidable uncertainties in bulk material and surface properties , verification of eq .
( 1 ) as a test of qed will likely always be inferior to measurements of the lamb shift or @xmath12 of the electron .
the experimental situation as of 1989 has been reviewed by sparnaay in the volume prepared in the honor of dr .
casimir s 80th birthday @xcite .
since then , two experiments have been performed , both with significantly better accuracy than had been previously obtained .
these two experiments were based on a torsion pendulum balance @xcite and on atomic force microscopy ( afm ) @xcite .
the recent experiments employed techniques that were developed in particular by van blokland and overbeek @xcite in the measurement of the attractive forces between metallic films .
measurements between metallic films pose difficult problems as compared to dielectric films for which optical techniques can be used for alignment and distance measurements . in the case of metallic films ,
the distance is determined by measurement of the capacitance between the plates .
alignment is simplified by making one plate convex , in which case the geometry is fully determined by the radius of curvature at the point of closest approach , and the distance between the plates at that point .
this technique was first put forward by deryagin @xcite and has found broader application as the proximity force theorem @xcite . for this geometry ,
( 1 ) becomes @xmath13 where @xmath14 is the radius of curvature and @xmath15 is the energy per unit area that leads to the force in eq .
it should be noted that the plate area does not enter into eq .
the two recent experiments each used one convex and one flat plate . because i was directly involved in this experiment
, i will give an anecdotal account of it ; specific details , schematics , and data can be found in @xcite .
this experiment was started in 1994 as an undergraduate project .
dev sen , a junior at the university of washington at that time , was working with our atomic physics group ; he heard of the casimir force in his electromagnetism course and asked if we might be able to measure it . with no knowledge of previous experimental work , with the exception of the sparnaay paper of 1958 @xcite ( which is routinely quoted as `` the '' verification of eq .
( 1 ) ) , i proposed an electrostatically balanced hanging ( cavendish ) torsion pendulum technique ; this is different from the horizontal torsion balance used by van silfhout @xcite ( which , at the time , we did not know about ) particularly because the angular force constant @xmath16 is at least two orders of magnitude smaller for the hanging pendulum . for short times
( a few seconds of averaging ) our principal noise source was thermal fluctuations ( e.g. , brownian motion ) ; in this limit , the signal to noise scales as @xmath17 .
it appears that this noise was a major limitation of our experiment , particularly when measuring the force at large plate separation .
we constructed the apparatus with junk found around the lab , and the total material construction cost of the apparatus was around $ 300 .
the torsion pendulum fiber was about 60 cm in length ; mounted on one arm of the pendulum was a flat plate coated with a cu / au film , while the other arm served as one electrode of a differential capacitor system that was used in a feedback system to keep the torsion pendulum at a fixed angle .
the second plate used for the casimir force measurement was mounted on a precision mechanical mount and could be moved about 10 @xmath18 m by use of piezoelectric transducers . in our first attempts to measure the casimir force ,
two flat plates were used .
we were never able to properly align the plates , and we eventually gave up when dev graduated .
some months before i moved to los alamos , i decided to give the apparatus one last try .
i had the idea to replace one flat plate with a convex plate ; i also improved the vacuum and electronic circuitry .
in june of 1996 the apparatus became operational ; when the two plates were moved into close proximity , the increase in force over the residual electrostatic interaction was easily evident .
unfortunately , the apparatus was not stable enough to make a measurement ; about 20 minutes after adjusting the distance between the plates so that the separation at closest approach would be of order 1 @xmath18 m or less , the apparatus would drift until the closest approach would increase to over 5 @xmath18 m .
i eventually discovered the source of the drift : the concrete floor of the physics department basement would distort while i was standing near the apparatus during adjustment , and the floor would relax to its unstressed position over a 20 minute period .
this effect was dealt with by standing in a different position , rather far from the apparatus , while making the initial adjustments ; the experiment thereafter became physically ( in an anatomical sense ) painful .
data was taken over a one month period , after which the apparatus was calibrated .
the final measurement was the radius of curvature of the cu / au coated convex plate ( lens ) in which i made an error . to make this measurement , i used
the angular deviation of a laser beam reflected from the lens surface as the lens was translated normally to the beam .
measurements near the center yielded @xmath19 cm , while translations across the entire lens yielded @xmath20 cm ; it did not occur to me that the lens might be aspheric ( it was a high quality , precision optical lens of unknown origin or application that was coated with a thin layer of au on top of a relatively thick cu layer ) .
i assumed the latter number was more accurate , and the data was described by eq .
( 2 ) with no corrections for finite conductivity ; i mistakenly convinced myself that the corrections should be less than 5% . at this point ,
my work was rudely interrupted by my relocating to los alamos , and i published my results in physical review letters .
subsequently , a number of theorists expressed surprise that no correction due to finite conductivity was required @xcite , and several expressed an interest in doing a proper calculation .
none did the calculation , so i figured out how to do it myself@xcite ; i knew that the plasma model correction to first order was not applicable ( see sec .
iiia ) ; being upwards of a 30% correction , none of the uncertainties in calibration were large enough to encompass it .
however , the variation in the measurement of @xmath14 was in the back of my mind , and in october of 1997 i made a trip to seattle and retrieved the lens . using a mechanical gauge , i measured @xmath21 cm in the region where the casimir force measurement was made .
the accurate finite conductivity correction is small enough that its effect is indistinguishable from a calibration error , or a mismeasurement of @xmath14 . in the end
, my data did not agree with the assumption that the film was pure au , but was better described by a pure cu film @xcite .
this result is not so surprising considering that the properties of the film depend on preparation technique , purity , and the possibility that the au diffused into the cu layer significantly
. a more accurate calculation of the finite conductivity effect would require a direct measurement of the complex permittivity of the films .
in hindsight , it is remarkable that the torsion balance experiment , which was intended as a demonstration , worked as well as it did .
the improvement over previous measurements is due to a number of factors , including the high sensitivity of the hanging torsion pendulum and its lack of mechanical hysteresis , larger measurement distances so vibration and mechanical instabilities were less important , improved piezoelectric transducers , and automated data collection so that large amounts of data could be analyzed and averaged .
much improvement over the present accuracy obtained by this technique is unlikely .
the apparatus was rather unwieldy with its enormous vacuum can and its susceptibility to tilt .
the length of the torsion fiber might be significantly shortened , reducing both the intrinsic sensitivity ( bad ) and sensitivity to external perturbations ( good ) ; however we must bear in mind that a factor of ten improvement in sensitivity only extends the measurement distance by factor of about two . in his 1989 review @xcite
, sparnaay discusses the possibility of using atomic force microscopy ( afm ) to measure the casimir force ; afm had just been invented at that time @xcite .
it was not until late 1998 that results from an afm casimir experiment were reported by mohideen and roy @xcite . in this experiment , an au / pd @xmath22 al coated
, 0.3 mm polystyrene sphere is attached to an afm cantilever .
a similarly coated optically polished sapphire plate was attached to a piezoelectric transducer and brought near the sphere .
the attractive force was determined by reflecting a laser beam from the cantilever tip ; the displacement of the laser beam on a pair of photodiodes produced a difference signal proportional to the cantilever bending angle .
the sensitivity of the apparatus was such that the absolute force could be determined with a fractional error of 1% at @xmath23 nm , and about 100% at 900 nm .
the use of afm to measure the casimir force might be a real breakthrough ; this is because the afm technique is very stable and reproducible .
unfortunately , it is limited to measurements short distance where there are significant theoretical uncertainties in the interpretation of the data .
for the first time testing eq .
( 1 ) or its modification for real materials to better than 1% accuracy appears possible .
as described in @xcite , it is anticipated that a factor of 1,000 improvement in sensitivity appears possible , which would extend the separation where the casimir force can be measured to 1% accuracy to about 1 @xmath18 m . at this distance ,
the theoretical uncertainties associated with the corrections for real materials , as described in the next section , become much less important .
equation ( 1 ) must break down when the plate separation is so small that the mode frequencies being affected when @xmath4 is varied are above the material resonance or plasma frequencies . in the case of a simple metal ,
the real part of the dielectric constant can be approximated by @xmath24 where @xmath25 is the the plasma frequency and is proportional to the effective free electron density in the metal .
it is convenient to introduce the plasma wavelength , @xmath26 .
corrections to eq .
( 1 ) , expanded in terms of @xmath27 , have been calculated to first order by haergraves @xcite and by schwinger et al .
@xcite , and to second order by bezerra et al.@xcite for flat plates , the corrected force can be written in terms of a eq .
( 1 ) with a multiplicative factor , @xmath28.\ ] ] this equation is only valid for @xmath29 ; unfortunately , the casimir force is large enough to be accurately measured experimentally only in the range @xmath30 or larger .
we are also faced with the problem that eq .
( [ plas ] ) is only approximate .
it is , however , possible to very accurately determine the attractive force as a function of plate separation by numerical calculation , provided we know its complex permittivity as a function of frequency : @xmath31 where @xmath32 are real . with this information , the permittivity along the imaginary axis
can then be determined by use of the kramers - kronig relation , @xmath33 this can be used in the lifshitz expression for the attractive force @xcite , @xmath34 ^ 2\over[s - p]^2}e^{2p\xi d / c}-1\right]^{-1}\nonumber\\ & + & \left[{[s+\epsilon(i\xi)p]^2\over [ s-\epsilon(i\xi p)]^2}e^{2p\xi d / c } -1\right]^{-1}\bigg ) dp\ d\xi\end{aligned}\ ] ] where @xmath35 . the numerical calculation for the attractive force between au , al , and
cu plates has been recently published @xcite , and significant deviations from eq . ( [ pcor ] ) were found . in particular , for al with @xmath36 nm eqs .
( [ pcor ] ) and ( [ lif ] ) differ by about 5% ; one should note that including the third order correction to eq .
( [ pcor ] ) worsens the deviation .
however , these calculations should be considered in light of the notorious variation of bulk and surface properties of materials due to preparation technique , purity , etc . @xcite . from the earliest experiments
, it was realized that surface roughness would lead to an increase of the apparent casimir force and therefore cause systematic errors in measurements aimed at verifying eq .
such effects were observed by van blokland and overbeek @xcite ; roughness has been discussed theoretically by van bree et al .
@xcite , and more recently in @xcite . for high - quality optically polished surfaces , the roughness rms amplitude @xmath1 is usually of order @xmath37 nm or less . for a @xmath9 attractive force , the correction to eq .
( 1 ) can be written @xmath38.\ ] ] the correction for the recent torsion balance experiment , at the point of closest approach , is about 1% , while for the afm experiment , it is about 30% .
the roughness correction was derived in the context of a @xmath9 force law ( this can be easily modified for the spherical plate @xmath8 case ) . however , the finite conductivity correction , particularly as given by eq .
( [ pcor ] ) , effectively has terms containing @xmath39 and @xmath40 . in principle , the roughness correction should be done for each power law separately , or the average force determined from the accurate calculation , eq .
( [ lif ] ) .
one should also bear in mind that the simple geometrical averaging procedure is nt exactly correct ; a complete treatment would involved solving the appropriate electromagnetic rough boundary problem .
however , the geometrical averaging is correct so long as the period of the roughness is larger than the separation between the plates .
either intentionally ( au evaporated onto an al or cu coated substrate ) or accidentally ( formation of oxide layers ) every casimir force measurement has made use of mono- or multilayer coated plates .
the calculation of the force for a general film configuration has been given by spruch and zhou @xcite .
the problem that i will consider here is very simple so i will outline its solution in some detail . a simple geometry that illustrates the effect of a thin material film is shown in fig .
1 ; one of two identical perfectly conducting flat plates is coated with a thin layer ( thickness @xmath41 ) of a real substance ( au , for example ) , and the separation between the perfectly conducting surfaces is @xmath42 .
this simplified problem will allow us to determine the qualitative effect of a thin film .
milonni has presented a complete calculation for the case of materials with no absorption ( @xcite ) , and this can be easily adapted to the multilayer case ( this calculational technique was first described in @xcite and elaborated upon in @xcite ) .
following milonni , we consider the case where the electromagnetic wave propagation vector in the three materials ( vacuum , film , perfect conductor ) @xmath43 where @xmath44 is a real number , and @xmath45 with 0 ( @xmath46 ) representing the space between the plates and 2 ( @xmath47 ) the perfect conductor , and we require @xmath48 .
for the case considered here , we must allow for a complex @xmath49 in which case the @xmath50s can be complex . in the appendix , the use of the techniques described by milonni for the case of absorption ( complex @xmath51 )
is justified .
there are two type of solutions to the wave equation , one with electric vector parallel to the surfaces ( with arbitrary orientation which we choose as the @xmath52 axis ) , @xmath53 and one with electric vector perpendicular to the surfaces ( along the @xmath54 axis ) , @xmath55 .
the wave equation is @xmath56 and the boundary condition for @xmath57 are ( 1 ) @xmath58 and @xmath59 are continuous , while for @xmath60 they are ( 2 ) @xmath61 and @xmath60 are continuous ( at the conducting surfaces , @xmath62 and @xmath63 ) . ignoring unphysical exponentially growing solutions
, we have @xmath64 where the @xmath65 sets @xmath62 or @xmath63 at the conducting boundary located at @xmath66 .
we therefore have two sets of linear equations involving @xmath67 for the two cases .
the condition for non - trivial solutions of these equations is that the determinant of the coefficient matrix is zero , yielding the following two expressions : @xmath68\over [ ( e^{2k_1a}+1)k_1-(e^{2k_1a}-1)k_0]}e^{2k_0d}-1\end{aligned}\ ] ] @xmath69\over [ ( 1-e^{2k_1a})k_1+\epsilon_1(\omega)(e^{2k_1a}+1)k_0]}e^{2k_0d}-1.\end{aligned}\ ] ] the zeroes @xmath70 of @xmath71 determine the allowed mode `` eigen '' frequencies , the zero point energy associated with the plates is determined by assigning energy @xmath72 to each mode ; @xmath73\ ] ] ( in general , the eigen - frequencies are complex , but the imaginary parts cancel as discussed in the appendix ) . where the sum over @xmath44 , in the continuum limit , becomes an integral , @xmath74 where @xmath75 is the transverse dimensions of the plate in the @xmath76 directions .
the theory of complex function can be used to evaluate the sum over eigen - frequencies ; specifically , according to the argument theorem @xcite , @xmath77 where @xmath78 is a closed path in the complex plane , @xmath79 and @xmath80 are the number of zeros and poles within @xmath78 , respectively , and the path is counterclockwise .
the argument theorem can be modified to give the sum of the zeroes and poles : @xmath81_{f(z_i)=0}- \left[\sum z_i\right]_{f(z_i)=\infty}.\ ] ] furthermore , @xmath82 .
the eigen - frequencies of physical interest lie in the right half plane ; integrating along the imaginary axis from @xmath83 to @xmath84 and closing the path with a semi - circle at infinity around the right half plane ( see @xcite for details ) , and integrating by parts gives @xmath85\ ] ] where we have set @xmath86 , with @xmath87 real , @xmath88 , and @xmath89 finally , we note that the poles of eqs .
( [ 1 ] ) and ( [ 2 ] ) do not depend on @xmath4 because it only enters in the multiplicative exponential ; therefore , eq . ( [ energy ] ) gives the zero point energy up to an additive constant , while the force per unit area is given by @xmath90\ ] ] where ( possibly ) non - physical @xmath4-independent terms are omitted .
@xmath91 for a au film 35 nm thick is shown in fig . 2 ;
@xmath92 was determined from tabulated optical constants as described in @xcite .
in the last few years , two new experiments to verify the existence of the casimir force between solid bodies have been performed and have shown a new level of accuracy ; in particular , atomic force microscopy technique offers great promise for testing eq .
( 1 ) and its modifications for the case of real materials to better than 1% precision . in particular ,
if the sensitivity of the afm experiment can be increased substantially , the measurement region can be extended to larger separations where the theoretical uncertainties discussed above are substantially reduced .
perhaps even the effect of finite temperature @xcite will be measurable in the not - too - distant future .
as discussed by milonni @xcite , the contour integration technique used to sum the eigen - frequencies is technically correct only if the eigen - frequencies lie on the positive real axis .
there has been much commentary on the extension of this technique to absorptive materials ( i.e. , complex permittivity ) in which case the eigen - frequencies are complex .
barash and giszburg @xcite introduce the idea of an auxiliary system to account for the complex permittivity , with the fundamental eigen - frequencies real .
however , the contour integration method gives the correct answer quite simply ; the mathematics does not know about the auxiliary system .
i would like to offer a non - rigorous explanation for why the technique works in the case of absorption .
first , for a generalized permittivity @xcite , @xmath93 .
therefore , the eigen - frequencies for the general boundary problem case occur in pairs , @xmath94 .
we further take the case where @xmath95 are either all positive or all negative which follows from continuation ; as @xmath96 , all the @xmath50 become equal because @xmath97 . in the case
that @xmath98 , representing exponentially damped surface wave , the eigen - frequencies lie in the lower half plane ; therefore , @xmath99 is damped exponentially in time .
for @xmath100 , the eigen - frequencies lie in the upper half plane and represent solutions growing exponentially in time and space .
clearly , the contour integration method as described above should not work without justification regarding branch cuts etc .
the way to see around this problem is instead of considering the eigen - frequencies , one can consider the corresponding eigen - wavenumbers @xmath50 , and @xmath101 in particular .
the eigen - wavenumbers occur in complex conjugate pairs @xmath102 ( as can be seen from eq .
( [ vec ] ) and the properties of @xmath103 in the complex plane ) and by definition the @xmath50 are in the right half plane for the exponentially damped solutions .
furthermore , if we write the determinant function in terms of @xmath101 , and using the fact that @xmath104 , we find @xmath105 and we can replace @xmath106 by @xmath107 because the path is arbitrary in the complex plane ( note that in the case of no absorption , the eigenvalues for @xmath101 lie on the imaginary axis while those for @xmath108 lie on the real axis , and the contour must be adjusted accordingly ) . because the eigenvalues for @xmath101 occur in conjugate pairs , the left side of eq .
( [ wcont ] ) is real ; therefore , the sum in eq .
( [ zsum ] ) is real , as can be seen from eq .
( [ csum ] ) when the integral is taken along the imaginary @xmath108 axis .
a. sarlemijn adn m.j .
sparnaay , _ physics in the making : essays on developments in 20th century physics in honour of h.b.g .
casimir on the occasion of his 80th birthday(north - holland , amsterdam , 1989 ) .
235 - 246 . _ | a brief review of the recent experimental verifications of the casimir force between extended bodies is presented . with modern techniques
, it now appears feasible to test the force law with precision better than 1% ; i will address the issues relating to the theoretical interpretation of experiments at this level of accuracy .
keywords : casimir force , quantum electrodynamics , fluctuations pacs numbers : 12.20.fv,07.07mp ( submitted for the casimir 90th birthday issue of comments on atomic and molecular physics ) |
a decade has passed since the discovery of the association between the long - duration gamma ray burst ( grb ) 980425 and the broad - lined type ic sn 1998bw @xcite .
this realization led to a renewed interest in the study of stripped - envelope core - collapse supernovae ( sne ) . in the case of sn 1998bw , the high kinetic energy inferred from the expansion velocity coupled with an exceedingly high luminosity and strong radio emission set it apart from all other previously - observed core - collapse sne .
since then these attributes have been shown also by other grb - related sne like sn 2003lw @xcite and sn 2003dh @xcite .
the x - ray flash sn 2006aj @xcite , was not as extreme as the previous sne , nevertheless it shows expansion velocity much higher then normal type ic sne .
in addition to these grb related , broad - lined sne ic , there have also been a number of other discovered broad - lined events that are apparently not associated with a grb , e.g. sn 1997ef , @xcite , sn 2002ap @xcite , sn 2003jd @xcite and sn 2007ru @xcite .
recently two other broad - lined events have been published that underscore the heterogeneous nature of this family of sne . in particular ,
early phase spectra of the broad - lined type ic sn 2007bg exhibit evidence of _ helium _
@xcite , while @xcite have presented the first case of a _ hydrogen - rich _ broad - lined type iib sn 2003bg .
in this paper we present ultraviolet , optical and near - infrared photometry and optical spectroscopy of sn 2009bb .
this object was discovered by the chilean automated supernova search chase @xcite on 2009 march 21.11 ut with the panchromatic robotic optical monitoring and polarimetry telescope ( prompt ) 3 at the cerro tololo inter - american observatory ( ctio ) .
the sn is located ( see figure [ fig1.1 ] ) at @xmath10 @xmath8 @xmath11 and @xmath12 @xmath8 @xmath13 ( equinox j2000 ) @xcite , which is about 170 west and 135 south of the center of the host galaxy ngc 3278 .
sn 2009bb was not visible in an unfiltered chase image ( mag @xmath14 19.2 ) obtained two days prior to the discovery image on 2009 march 19.20 ut . we were therefore able to use this image in combination with the first follow - up images to obtain a tight constraint on the explosion epoch ( see section 2.3 ) .
as this was an extremely young sn , an intensive follow - up campaign was organized using the prompt telescopes @xcite and the facilities available to the carnagie supernova project ( csp ; hamuy et al .
2006 ) at las campanas observatory ( lco ) .
a week after discovery , @xcite classified sn 2009bb as a broad - lined sn ic .
radio and x - ray observations were obtained with the vla and chandra telescope .
an analysis of these data indicates that sn 2009bb was associated with strong radio emission and mild x - ray emission @xcite .
these findings are suggestive of the presence of a relativistic outflow typically associated with grb - related sne .
the explosion site of sn 2009bb has been studied in detail by @xcite who showed that contrary to other jet - driven sn explosions , which to date have always been identified with metal poor environments [ see @xcite for a review ] , the explosion site of sn 2009bb exhibits a metallicity between 1.7 to 3.5 z@xmath4 .
@xcite also found a super - solar metallicity for the environment of the long - duration grb 020819 .
these findings present a challenge to the theoretical framework developed to explain grbs @xcite .
however , @xcite have recently proposed a model in which the progenitor star does not need to be in a low metallicity environment . in this scenario
sn 2009bb represents a peculiar object that could open new theoretical developments .
the organization of this article is as follows .
the photometric data are analyzed in section 2 , and in section 3 the pseudo - bolometric light curve is used to estimate some physical parameters of sn 2009bb .
an analysis of the spectroscopic evolution of sn 2009bb is carried out in section 4 , and section 5 presents a discussion and summary of the major results .
observation and data reduction techniques are described in appendix a.1 and a.2 .
the @xmath15 photometry of sn 2009bb is reported in table 1 , while the light curves are shown in figure [ fig2.1 ] .
also plotted in the latter figure are photometry of the broad - lined type ic sne 1998bw , 2002ap , 2003jd , 2006aj and sn 2007bg . from this comparison of light curves ,
it is clear that in the @xmath16 bands during the pre - maximum phase , sn 2009bb , sn 2006aj and sn 2003jd show a similar evolution , while sn 1998bw and sn 2002ap have a slower and faster rise - time , respectively . in the @xmath2 band ,
sn 2009bb shows , with the exception of sn 1998bw , a slower rise than all the other sne included in the plot . at post - maximum phases ( epoch @xmath17 @xmath120 days ) , the @xmath16 light curves of sn 2009bb decrease in magnitude
faster than the other objects , except for sn 2006aj , which appears to be quite similar to sn 2009bb .
this is not the case in the @xmath2 band where the sn 2006aj decline rate is clearly faster than that of sn 2009bb . in the latter band , the object that most resembles sn
2009bb is sn 2003jd , while sn 2002ap and sn 1998bw display much slower decline rates .
the @xmath18 photometry of sn 2009bb is reported in table 2 , while the light curves are shown in figure [ fig2.2 ] .
these are the first - ever observations of a broad - lined sn ic in the sloan bands . for comparison in the plot
, we also include @xmath19-band light curves of sn 1998bw , sn 2002ap and sn 2006aj .
the @xmath20 light curve of sn 2009bb most resembles the @xmath19-band observations of sn 2002ap , but again declines more rapidly than sn 1998bw .
it should be stressed that figure [ fig2.2 ] compares light curves on different photometric systems , and therefore the comparison should be taken with caution .
the @xmath21 photometry of sn 2009bb is reported in table 3 . in figure
[ fig2.3 ] , the near - infrared light curves of our sn are displayed together with those of sn 1998bw and sn 2002ap .
as seen in the optical , the post - maximum evolution of sn 2009bb is slightly faster than that of sn 2002ap . in the case of sn 1998bw , the difference is larger , with the latter object being @xmath30.7 and @xmath30.2 magnitudes brighter than sn 2009bb in the @xmath22 and @xmath23 bands , respectively , at @xmath2450 days past @xmath2 maximum .
is is worth mentioning that part of this difference could be due to host galaxy contamination affecting the sn 1998bw infrared photometry @xcite . for each optical and near - infrared band , the time and value of peak magnitude was estimated from low - order polynomial fits .
the results are reported in table 4 .
similar to other sne ib / c , maximum light occurs in the blue followed by the red bands . in figure [ fig2.0 ] , the count rates in the @xmath25 and @xmath26 uvot bands at the position of sn 2009bb are plotted .
the fluxes were measured in a fixed aperture and , therefore , are the sum of the light coming from the sn and the background galaxy . from the plot ,
it is clear that in the @xmath25 filter , the background flux is already dominant at @xmath24 6 days past @xmath2 maximum light . assuming that the count rate measured in the last three @xmath25 observations is entirely due to the background , the sn 2009bb flux in the first observation can be estimated by subtracting the weighted average of the flux in these last three observations from the total count rate .
adding a zero point for the @xmath25 band of 17.49 @xmath3 0.03 @xcite , a value of 18.1 @xmath3 0.1 is obtained for the @xmath25 magnitude of sn 2009bb at @xmath24 6 days before @xmath2 maximum light .
the de - reddened color curves of sn 2009bb are compared in figure [ fig2.4 ] with those of sn 1998bw , sn 2002ap , sn 2006aj , and sn 2003jd . in the case of sn 2009bb , a color excess of e(@xmath27 ) = 0.58@xmath28 is assumed ( see section 3 ) , while for sn 1998bw , sn 2002ap , sn 2003jd , and sn 2006aj , values of 0.06 @xcite , 0.08 @xcite , 0.14 @xcite , and 0.18 @xcite , respectively , were adopted .
it is worth mentioning that for all the latter sne most of the reddening is produced in the milky way .
all the authors cited above assumed little or no reddening in the host galaxy due to the absence of strong interstellar na i d absorption in the sne spectra . since the correlation between na i d @xmath29 and reddening has been shown to have a significant dispersion @xcite , the reddening estimates for these sne should be taken with caution . to convert the color excesses to corrections for each band
, we adopted an @xmath30 value of 3.1 @xcite . as is seen in figure [ fig2.4 ] ,
the color evolution of broad - lined sne ic show significant variations .
this highlights the inadequacy of color indices to estimate host galaxy extinction for these events .
the early phase @xmath31 color evolution of sn 2009bb is seen to be most comparable to that of sn 2003jd , while sn 1998bw is @xmath240.2 mag redder at early and late epochs , but is similar to sn 2009bb between + 10 and + 20 days . in the case of @xmath32 ,
all sne have very similar evolution except for sn 2006aj which is @xmath240.2 redder .
the @xmath33 color evolution of sn 2009bb is intermediate between sn 2002ap ( bluer ) and sn 2003jd ( redder ) . in figure [ fig2.5 ]
, we compare the reddening - corrected @xmath34 and @xmath35 color curves of sn 2009bb with those of sn 2002ap .
the evolution of the two sne is similar , but sn 2009bb is @xmath240.3 and @xmath240.5 mag bluer in @xmath34 and @xmath35 , respectively .
the short time interval between the chase images with negative and positive detection of sn 2009bb allows us to place strong constraints on its explosion date .
this is done through the use of a simple equation describing an expanding fireball : @xmath36 where @xmath37 is the luminosity at time , @xmath38 , @xmath39 is a constant that defines the rate of the rise , n=1.8 @xcite , and @xmath40 is the time of explosion . applying the equation ( [ equa6 ] ) to the unfiltered light curve
, we estimate that the explosion occurred on jd=2454909.6 @xmath3 0.6 , which corresponds to the time of of our non - detection of sn 2009bb .
as demonstrated in figure [ fig2.6 ] , the curve fits reasonably well the unfiltered light curve ( reduced @xmath41=3.1 ) , except for the epoch of first detection .
to compute the absolute luminosity of sn 2009bb requires estimates of the host galaxy dust extinction and distance .
one method to estimate the host reddening of sn 2009bb is to measure the balmer decrement from the spectrum of the region located beneath the sn . to this end
we use the two sn 2009bb nebular spectra acquired on jan 9th and feb 3th 2010 . in these spectra
the signal - to - noise ratio of the region is higher than in the spectra taken at earlier epochs , while the sn flux is lower , allowing for a more precise measurement of emission from the balmer lines .
obtaining an accurate estimate of the h@xmath42 and h@xmath43 fluxes requires the removal of the light of the sn and any residual stellar contribution .
this was accomplished by fitting a cubic spline to the underlying continuum in the spectrum .
the resulting pseudo - continuum was then subtracted from the original spectrum . comparing the measured @xmath44 flux ratio obtained from
the `` cleaned '' spectra to the values listed in @xcite ( for case b recombination , t=10000@xmath45 k ) and assuming the standard total - to - selective extinction ratio r@xmath46 = 3.1 , we obtain a total reddening ( host+milky way ) of @xmath47 = [email protected] and @xmath47 = [email protected] for the 2010 jan 9 and 2010 feb 3 spectra , respectively .
there is another ( brighter ) region located @xmath24 0.3@xmath48 north , @xmath24 1.4@xmath48 east of the sn .
a measurement of the balmer decrement from its spectrum gives a color excess of @xmath47 = [email protected] , which is in good agreement with that obtained at the sn position .
given the similarity of the three estimates , we will assume the total reddening suffered by sn 2009bb to be the average of the values obtained from its two nebular spectra , i.e. @[email protected] . an alternative method often used to estimate the reddening suffered along the line - of - sight to a sn is the equivalent width ( @xmath29 ) of the interstellar na i d absorption .
the weighted average of the @xmath29 measured from seven of our low resolution spectra is @xmath49 .
taking into account relation between the na i d ew and color excess reported in @xcite , a value of @xmath50 and @xmath51 is implied for the lower and higher slope , respectively .
this confirms that sn 2009bb was significantly extinguished .
nevertheless , since the na i d lines are almost certainly saturated , this value of @xmath52 should be taken as a lower limit only . combining the galactic reddening obtained from the infrared dust maps of @xcite , @xmath530.098 , and the host galaxy value , we obtain a lower limit of the total reddening of @xmath54 and @xmath55 , respectively , bracketing the reddening obtained from the balmer decrement .
despite the relative proximity of ngc 3278 , there are no direct distance measurements to this galaxy reported in the literature .
we therefore must resort to the recession velocity and hubble s law to compute the distance to sn 2009bb . in the remainder of this paper
we assume a distance modulus of @xmath56 as given in the nasa / ipa extragalactic database ( ned ) .
this value is obtained from a heliocentric velocity of 2961@xmath337 km s@xmath57 @xcite , corrected for virgo infall and great attractor @xcite , and assuming @xmath58 km s@xmath57 mpc@xmath57 @xcite . using the maximum light magnitudes reported in table 4 , and the previously mentioned values of reddening and distance
, we compute absolute magnitudes for the various bands that are listed in table 4 .
figure [ fig3.1 ] shows the pseudo - bolometric light curve of sn 2009bb , together with sn 1998bw , sn 2002ap , sn 2003jd and sn 2006aj , obtained by integrating the absolute flux in the @xmath15 bands .
the peak luminosity of sn 2009bb is found to be close to that of sn 2006aj , and @xmath2460% of that of sn 1998bw . to estimate the @xmath5ni content and
the ejected mass , we integrated the flux of sn 2009bb contained within the @xmath59 bands . also included was a @xmath39 band correction based on the sn 2002ap light curves @xcite .
the resulting pseudo - bolometric light curve was then modeled making use of an analytical description for the peak of the light curve @xcite .
the model assumes spherical symmetry , homologous expansion , no mixing of @xmath5ni , radiation - pressure dominance , and the applicability of the diffusion approximation for photons , which restricts it to early phases ( @xmath14 30 days post - explosion ) when the density is sufficiently high to make the ejecta optically thick .
we also assume constant opacity @xmath60= 0.08 @xmath61 g@xmath57 as computed for sn 1997ef by @xcite .
@xcite used a time variable opacity for modelling the sn 1998bw bolometric light curve which average value during the first 20 days after the explosion is @xmath60= 0.07 @xmath61 g@xmath57 .
we assume the uncertainty on visible opacity to be the difference of the previously mentioned values i.e. 0.01 @xmath61 g@xmath57 . in @xcite model ,
the time - evolution of luminosity is given by @xmath62 where @xmath63 is the energy produced in 1 second by 1 gram of @xmath5ni , @xmath64=t/@xmath65 and @xmath66=@xmath65/2@xmath67 with @xmath67 being the @xmath68-folding time of the @xmath5ni decay and @xmath65 effective diffusion time , which determines the width of the bolometric light curve .
if @xmath65 is expressed as a function of @xmath60 , the ejecta mass , @xmath69 , and the photospheric velocity , @xmath70 at time of bolometric maximum , the effective diffusion time can be written @xcite as @xmath71 in addition to the m@xmath72 and m@xmath73 parameters , there is also an additional free parameter , t@xmath74 , which is the time interval between explosion and the peak of the bolometric light curve .
finally , as proposed by @xcite , we include the contribution of the @xmath5co decay in equation [ equa3 ] . in this analytic model ,
the photospheric velocity at the time of the maximum of the bolometric light curve is an input parameter .
the velocity derived from the minimum of the @xmath756355 feature is often used as a proxy for the photospheric velocity .
nevertheless , several studies have found that in sn ic the @xmath756355 could be contaminated by other species like detached @xmath756678 , @xmath756402 , detached h@xmath42 or detached @xmath756580 @xcite .
contamination from other ions could shift the minimum of the @xmath756355 feature in different direction at different epochs , biasing the estimation of the photospheric velocity . in spite of these shortcomings
the @xmath756355 feature is still the most easy feature to measure , but a correction must be applied to derive a velocity as close as possible to the real photospheric velocity at a given epoch . to compute this correction we consider sn 1998bw , sn 2002ap and sn 2006aj , for which the evolution of the photospheric velocity was computed using the montecarlo code described in @xcite . comparing the velocities measured from the minimum of the @xmath756355 absorption with those derived through the spectra modelling mentioned above ,
a difference of @xmath243000 km s@xmath57 is found .
assuming that this velocity shift also applies to sn 2009bb , we derive @xmath76 @xmath8 15000 @xmath3 1000 km s@xmath57 at the time of the bolometric maximum light ( jd=2454923.1 @xmath3 0.7 ) .
the analytical model then yields m@xmath73 = 0.22 @xmath3 0.06 m@xmath4 , m@xmath72 = 4.1 @xmath3 1.9 m@xmath4 and t@xmath74=11.3 @xmath3 0.8 days .
the uncertainties in the latter quantities were computed through montecarlo simulations , taking into account the errors in the photospheric velocity , optical opacity , distance modulus , reddening and photometry .
the resulting model is compared to the computed bolometric light curve in the insert of figure [ fig3.1 ] .
the ejecta mass and photospheric velocity allow us to estimate the kinetic energy associated with the 2009bb explosion by comparing the pseudo - bolometric light curve of sn 2009bb with that of sn 2002ap , for which detailed modelling have been already carried out @xcite . to compute the kinetic energy , equation [ equa4 ]
can be written as follows : @xmath77 around maximum the pseudo - bolometric light curves of sn 2002ap and sn 2009bb are similar ( see the rescaled pseudo - bolometric light curve of sn 2002ap in the insert of figure [ fig3.1 ] ) .
the decline rate of sn 2002ap is slightly slower than sn 2009bb .
around two weeks after @xmath2 maximum , which is the latest time we considered in the bolometric light curve fit , sn 2002ap reaches the same brightness of sn 2009bb @xmath24 3 days later than the latter sn .
nevertheless , the rise to maximum of sn 2002ap seems to be faster than that of sn 2009bb .
computing the rise time of the bolometric light curve as difference between the explosion date and the epoch on which the bolometric light curve reach the maximum , we obtain a rise time of @xmath24 13.5 and @xmath24 10 days for sn 2009bb and sn 2002ap respectively .
we therefore can assume @xmath78 @xmath24 @xmath79 .
taking into account that for sn 2002ap @xmath80 and @xmath81 = 4 @xmath3 1 @xmath82erg @xcite , for sn 2009bb we obtain @xmath83erg .
radio observations of sn 2009bb spanning @xmath84 days after the explosion revealed evidence for an extraordinarily luminous , long - wavelength counterpart , with a spectral luminosity @xmath85 at @xmath86 ghz of @xmath87 @xcite . in comparison to the other @xmath88
nearby sne ibc observed in the radio on a similar timescale , sn 2009bb is a factor of 10 to @xmath89 more luminous , rivaling the radio afterglow luminosities of the nearest grb - sne , including grb980425 associated with sn 1998bw @xcite .
the multi - frequency radio observations were well described by a self - absorbed synchrotron spectrum , produced as the blast - wave shock - accelerated electrons in the local circumstellar medium . just as in the case of sn 1998bw , the high brightness temperature of the sn 2009bb radio counterpart required an unusually large emitting region and , in turn , a trans - relativistic blast - wave velocity , @xmath90 .
moreover , the high radio luminosity implied a copious energy coupled to the fastest ejecta , @xmath91 erg .
@xcite appealed to a model where this trans - relativistic ejecta component was powered by a different mechanism , a `` central engine '' ( rapidly - rotating and accreting black hole or magnetar ) , as is commonly assumed for long - duration grbs @xcite .
this model can be tested using the bulk explosion parameters ( @xmath92 , @xmath93 ) for sn 2009bb derived from our well - sampled optical photometry and spectroscopy .
analytic solutions for the distribution of ejecta in type ibc sne predict a steep coupling of energy and velocity within the homologously - expanding material , @xmath94 @xcite . based on an analysis of the optical dataset ( see section 3 ) , we find bulk ejecta parameters of @xmath95 erg and @xmath96 . within the homologous outflow
, we therefore expect the ejecta traveling at @xmath97 to carry a kinetic energy of @xmath98 erg .
this is _ at least _
three orders of magnitude below the energy budget required to power the trans - relativistic radio emitting ejecta .
we conclude that the luminous synchrotron emission from sn2009bb can not be explained in the framework of a homologous sn explosion and another energy reservoir ( i.e. a central engine ) is required to explain the distribution of ejecta within this relativistic sn .
optical spectra of sn 2009bb were obtained on thirteen epochs spanning phases between @xmath02 and @xmath1285 days .
table 5 provides a log of these observations , and figure [ fig4.0 ] illustrates the spectral evolution . comparing the de - reddened spectra of sn 2009bb to similar epoch spectra of sn 1998bw , sn 1997ef , sn 2003jd and sn 2002ap , de - reddened using the color excess values mentioned in section 3
, one can note that at early phases sn 2009bb , sn 2003jd and sn 2006aj are characterized by a bluer continuum than sn 1997ef and sn 2002ap ( see figures [ fig4.1 ] and [ fig4.2 ] ) .
around two week past maximum , the continuum has become similar to that of all the other comparison sne . in terms of degree of broadening of the spectral features around maximum , the spectra of sn 2009bb are similar to those of sn 2006aj ( see figure [ fig4.1 ] and [ fig4.2 ] ) .
looking at the sn 2009bb spectra taken later than a week after @xmath2 maximum ( see figure [ fig4.3 ] , figure [ fig4.4 ] and [ fig4.5 ] ) , the degree of blending is still high in sn 2009bb being intermediate between sn 2003jd and sn 1998bw .
this slow spectroscopic evolution in spite of the high expansion velocity is a further indication that sn 2009bb had a massive envelope .
the expansion velocity of 2009bb decreases similarly to that of sn 1998bw and slower sn 2002ap ( see figure [ fig4.8 ] ) . in the following ,
the evolution of some features present in the spectra of sn 2009bb are analyzed in more detail using the supernova spectrum - synthesis code synow @xcite .
particular effort is devoted to determining if helium is present in the ejecta of sn 2009bb , as this is very important for constraining the nature of its progenitor and its evolutionary stage at the time of the explosion .
we explore qualitatively the presence of helium using synow , but it must be kept in mind that full spectral modeling , including non - thermal effects is necessary to study quantitatively the evolution and strength of the he lines @xcite . the clearest evidence of the presence of helium in the sn 2009bb ejecta is found in the first two spectra taken at @xmath02 and @xmath01 days ( see figure [ fig4.1 ] ) . a striking feature in these two spectra is the absorption around 5400 .
this line is not visible in the other sne spectra , and we suggest that it corresponds to @xmath75 5876 .
fitting the line profile with synow , we obtain a good match for a velocity in the line forming region of @xmath24 28000 km s@xmath57 and 27500 km s@xmath57 in the @xmath02 and @xmath01 days spectra , respectively .
a similar velocity is also obtained by fitting the absorption around 4800 with @xmath994924 , 5018 , 5169 and also provides a good match with @xmath756355 for the absorption located at @xmath245800 . in figure
[ fig4.1 ] , the expected positions of the @xmath756678 and @xmath757065 absorption are also marked . in both the @xmath02 and @xmath01 day spectra ,
features are present at this wavelength , but due to their weakness and to strong background contamination , should not be over - interpreted . in figure [ fig4.2 ]
we show the sn 2009bb spectrum taken at @xmath14 days .
the velocity obtained from synow fits to @xmath994924 , 5018 , 5169 and @xmath756355 is @xmath24 18000 km s@xmath57 .
sn 2009bb shows the closest resemblance to sn 2006aj , but differences are visible particularly around the @xmath756355 absorption and between 5000 and 6000 , where sn 2009bb clearly differs compared to the other sne . attributing this absorption to detached at @xmath2425000 km s@xmath57
, we obtain a reasonable fit of the observed feature ( see short dashed green line in figure [ fig4.6 ] ) .
the expected positions of the other features for the assumed detached velocity ( @xmath2425000 km s@xmath57 ) are indicated in figure [ fig4.2 ] .
indeed , an absorption corresponding to @xmath756678 appears to be present and helps to fit the boxy profile of the line ( see figure [ fig4.6 ] ) .
there is also a weak feature corresponding to the expected position of @xmath757065 , but contamination from @xmath756929 can not be excluded .
figure [ fig4.3 ] shows the spectrum of sn 2009bb taken a week after @xmath2-band maximum .
between 5000 and 5500 , sn 2002ap and sn 2003jd show an absorption apparently due to .
this feature is absent in the spectra of sn 2009bb , sn 1998bw and sn 1997ef , which is probably due to the broadness of the spectral features in the latter sne , that tends to wash out weak features .
nonetheless , the absence of a clear feature in the sn 2009bb makes less plausible the possibility that in the earlier sn 2009bb spectra the absorption around 5500 is due to . as shown in figure [ fig4.6 ] , with a detached velocity of @xmath24 23000 km s@xmath57 gives a good fit to the observed line profile between 5000 and 5500 .
figure [ fig4.3 ] also shows the spectrum of the broad lined type ic sn 2007bg , for which the lines ( the positions of the @xmath757065 , @xmath757235 lines are marked in the graph ) appear to be even stronger than observed in the spectrum of sn 2009bb .
unfortunately , due to the higher expansion velocity of the absorption features in sn 2009bb , the @xmath757065 line falls in the gap between the ccds , making its presence impossible to confirm .
figure [ fig4.4 ] shows a spectrum of sn 2009bb taken two weeks after @xmath2-band maximum .
the absorption feature around 5500 is still deeper in sn 2009bb than in the other two sne . for sn 2009bb , an equally good fit to this feature
is obtained with undetached ( @xmath24 14000 km s@xmath57 ) , or with @xmath755876 detached at @xmath24 18000 km s@xmath57 .
again , the expected positions of the other prominent lines are indicated in this figure , but clear signs of these features can not be seen in the sn 2009bb spectrum .
figure [ fig4.5 ] displays the spectra taken on days @xmath130 and @xmath144 .
the emission redward of the absorption remains stronger in sn 2009bb compared to the the other sne plotted in this figure , but the profile around the absorption is now not so different suggesting a reduced strength of the blue wing of the @xmath755876 detached component . in both the @xmath133 and @xmath144 spectra of sn 2009bb
, there also is no clear signature of the @xmath756678 and @xmath757065 features . it should be noted that in the rather noisy spectrum of sn 2007bg , the helium lines , if present , are not as strong as expected . in the case of helium - rich core - collapse sne
( sne ib ) , the features are stronger at these later epochs than at maximum ( e.g. * ? ? ?
a possible explanation for the lack of strengthening of the absorption features in the case of sn 2009bb ( and perhaps also for sn 2007bg ) is that , due to the higher expansion velocity of broad - lined sne ic together with a less massive helium shell , the mass of helium in the line forming region at these later epochs is likely much lower than in normal sne ib .
the evolution of the @xmath755876 velocity is summarized in the insert of figure [ fig4.8 ] . before maximum he is undetached from the main line forming region , while after maximum is detached .
this agrees with the finding of @xcite and @xcite for type ib sne .
the velocity of the detached he decrease with time .
the latter also agree with what obtained by @xcite .
the fractional line depth of the @xmath757774 line can , in principle , be used to infer the presence of helium using the formula given by @xcite @xmath100 where @xmath101 and @xmath102 represent the values of the continuum and of the line minimum at the wavelength of the minimum , respectively .
the latter authors proposed that in sne ib , due to the dilution of oxygen by helium , the @xmath757774 absorption should be weaker than in sne ic .
indeed , they measured an average fractional line depth of 0.27 @xmath3 0.11 and 0.38 @xmath3 0.09 for sne ib and sne ic , respectively .
@xcite also found that this difference is stronger around maximum than at late time .
unfortunately , in broad - lined sne ic , the depth of the @xmath757774 is difficult to measure at early epochs due to the widths of the other lines in this spectral region . in figure
[ fig4.7 ] , measurements at phases later than + 20 days are plotted . as for normal sn ibc
the fractional line depth for broad - lined sne ic increases with time .
interestingly , the two broad - lined sne ic which show evidence of helium lines , sn 2009bb and sn 2007bg , have an @xmath757774 line depth that is less than those that do not show any sign of helium . note that figure [ fig4.7 ] does not include sn 1998bw since , even at + 50 days , @xmath757774 is still strongly blended with the near - infrared triplet . in figure [ fig4.9 ]
we show a comparison between the nebular spectra of sn 2009bb , sn 1997ef , sn 1998bw , sn 2002ap and sn 2006aj . to study the line profiles of sn 2009bb , the strong emission lines from the underlying region must be removed .
this was achieved by subtracting the spectrum of the bright region located @xmath24 0.3@xmath48 north , @xmath24 1.4@xmath48 east of the sn position .
the latter spectrum was acquired with the same spectrograph , making more accurate the spectra matching procedure .
the spectrum of the bright region was rescaled to match the line fluxes of the region underlying the sn .
note that the line ratios for the two regions are similar to within 10% .
this , together with the very similar balmer decrement measured in the two regions ( see section 3 ) , gives confidence that the subtraction procedure has not significantly altered the sn spectrum .
the nebular spectrum of sn 2009bb is rather noisy , but several strong emission features can be recognized .
the emission in the blue , near 4500 , should be dominated by [ ] @xmath754570 .
the broad complex near 5200 is the result of a number of [ ] emission lines .
near 5900 , d emission is possibly detected .
the two strongest lines in the spectrum are [ ] @xmath75@xmath756300 , 6363 and ] @xmath75@xmath757291 , 7324 .
it is immediately clear that the calcium to oxygen ratio is much larger in sn 2009bb than in all the other comparison sne .
it is also evident , when compared to the spectra of sne 2003jd and 1998bw , that the profile of the [ ] @xmath75@xmath756300 , 6354 emission in sn 2009bb is more similar to that of sn 1998bw than that of sn 2003jd ( see figure [ fig4.10 ] ) .
based on spectral modelling , @xcite claimed that a double peaked [ ] @xmath996300 , 6354 profile as observed in sn 2003jd is produced by an aspherical expansion , but viewed off axis . assuming that this interpretation is correct [ but see also @xcite ] , in the case of sn 2009bb , the material ejected must have been oriented close to the line of sight as in the case of sn 1998bw .
the nebular spectrum of sn 2009bb can be modelled in order to derive some physical properties of the sn .
nebular spectroscopy reveals the conditions of the innermost part of the sn ejecta , which are not visible when the sn is bright because they lie too deep in regions that are optically thick . at late phases , on the other hand , the inner parts of the sn are transparent , and behave like a nebula .
the code we used computes the energy deposition from the decay of @xmath5ni to @xmath5co to @xmath5fe .
these decays emit @xmath9-rays and high - energy positrons , which can deposit their energy in the sn ejecta and heat the gas by collision excitation .
cooling then takes place via line emission .
our code was described by @xcite , and is based on the description of @xcite .
we model the spectrum of sn 2009bb using the simple , one - zone version of the nebular code .
this assumes that the sn ejecta are spherical , and bound by an outer velocity , which reflects not so much the extent of the ejecta in radius or velocity ( the two are equivalent in the homologously expanding sn ejecta ) , but rather that of the region which is effectively heated and can therefore emit lines .
this may be the result of decreasing density , distribution of @xmath5ni , or both .
our model requires an outer velocity of 5500 km s@xmath57 .
this velocity appears to be adequate to fit all lines .
this could be taken as an argument that sn 2009bb was not significantly aspherical . in models where more energy is released in a polar direction , leading to the synthesis of @xmath5ni and possibly to the production of a grb , the [ ]
lines are broader than the [ ] line if the event is viewed close to the polar axis @xcite .
otherwise , the [ ] lines are narrower than the [ ] line , which shows a characteristic double - peaked profile , as in sn 2003jd .
nevertheless , the strongest evidence for asphericity in sn 1998bw was seen in nebular spectra obtained 200 days after maximum @xcite , while the difference between the iron and oxygen lines width was smaller at epochs of 340 days , which is closer to the epoch of the only nebular spectrum available for sn 2009bb .
therefore , while from this spectrum only we do not infer major asphericities in sn 2009bb , we can not rule out that the signature of asphericity might have been seen in earlier nebular data .
the @xmath5ni mass which is required to reproduce the spectrum , assuming a distance modulus of 33.0 and a reddening @xmath103 , is 0.25 m@xmath4 .
this is in good agreement with the estimate of [email protected] m@xmath4 found from modelling the pseudo - bolometric light curve .
the mass of the elements used to generate the synthetic spectra are summarized in table 6 .
we have presented uv , optical and near - infrared photometry , and optical spectroscopy of the broad - lined type ic sn 2009bb . around maximum ,
the spectra of sn 2009bb were similar to those of sn 2003jd and sn 2006aj .
the expansion velocities of sn 2009bb are similar to these of sn 2006aj , but higher than those of sn 2003jd . at phases later than + 15 days , the broadness of the spectral features in sn 2009bb was intermediate between sn 1997ef and sn 1998bw .
the slow spectroscopic evolution indicates that sn 2009bb had a massive envelope , and this is confirmed by the pseudo - bolometric light curve analysis . with a simple model that includes both the @xmath5ni and @xmath5co decay , and making use of analytic equations presented in @xcite , we obtain a @xmath5ni and ejected mass of [email protected] m@xmath4 and 4.1 @xmath31.9 m@xmath4 , respectively .
the resulting kinetic energy is @xmath104 foe .
a similar @xmath5ni mass is obtained from the nebular spectrum modelling @xmath105 .
an absorption feature is identified in the @xmath106 and @xmath107 day spectra around 5400 that we attribute to @xmath755876 forming at the photosphere .
appears to be detached in the later spectra , with the absorption getting weaker with time .
the latter evolution is odd compared to what is normally observed in sne ib where helium lines become stronger after maximum .
the weakening of the helium absorption could be due to a combination of the fast expansion characteristic of broad - lined sne and a less massive helium shell than in `` normal '' sn ib . the comparison between the kinetic energy carried away by the homologous expanding ejecta at @xmath97 and the energy inferred by the radio observations points to the need for an extra source of energy provided by a central engine .
the modelling of the sn 2009bb nebular spectrum indicates that iron and oxygen lines can be reproduced with a synthetic spectrum where both elements have similar expansion velocities .
this suggests that sn 2009bb ejecta structure was not significantly aspherical .
nonetheless , considering the iron and oxygen lines evolution of sn 1998bw , we can not exclude that the signature of asphericity could have been detected in a sn 2009bb earlier nebular spectrum .
the analysis of the [ ] @xmath996300 , 6354 reveals that , if the explosion was indeed aspherical , the orientation should have been similar to that of sn 1998bw .
therefore , if a grb was produced during the sn2009bb explosion , it was below the threshold of the current generation of @xmath9-ray instruments . without a detected grb counterpart , sn
2009bb represents the first engine - driven , relativistic supernova ever discovered by its optical emission alone .
the presence of a relatively massive helium layer may have played a role on the failed grb detection , but a quantitative description of the helium shell is necessary to verify this hypothesis . a special thanks to the gemini - south staff for their efforts in obtaining data for our spectroscopic program .
g.p . acknowledges support by the proyecto fondecyt 11090421 and from comit mixto eso - gobierno de chile .
ackowledges support from fondecyt through grant 1060808 .
, m.h . and j.m .
acknowledge support from the millennium center for supernova science through grant p06 - 045-f funded by `` programa bicentenario de ciencia y tecnologa de conicyt '' , `` programa iniciativa cientfica milenio de mideplan '' .
g.p . , m.h . and j.m .
acknowledge partial support from centro de astrofsica fondap 15010003 and by fondecyt through grant 1060808 from the center of excellence in astrophysics and associated technologies ( pfb 06 ) .
the dark cosmology centre is funded by the danish nsf .
this material is based upon work supported by the national science foundation ( nsf ) under grant ast0306969 .
this paper is based on observations obtained at the gemini observatory , cerro pachon , chile ( gemini programs gs-2009a - q-17 and gs-2009a - q-43 ) .
we have 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 .
imaging of sn 2009bb was acquired with five different instruments in the optical , one in the near - infrared , and one in the ultraviolet .
a summary of the characteristics of the facilities used are listed below . *
the swift space telescope is equipped with the uvot camera that has 256@xmath108256 physical pixels , but after processing the final combined images has 2048@xmath1082048 pixels with a pixel scale of 05 .
sn 2009bb was observed with the @xmath25 and @xmath26 filters . *
the prompt 3 telescope at ctio is equipped with a ccd camera alta u47uv e2v ccd47 - 10 ( 1024@xmath1081024 , pixel scale = 06 per pixel ) .
sn 2009bb was observed with both a standard johnson - kron - cousins @xmath2-band and a sloan @xmath109 filters . *
prompt 5 is equipped with a ccd camera alta u47 e2v ccd47 - 10 ( 1024@xmath1081024 , pixel scale = 06 per pixel ) .
sn 2009bb was observed with both johnson - kron - cousins @xmath16 and sloan @xmath110 filters . * the 1.0 m henrietta swope telescope at lco
is equipped with a site3 ccd camera ( 2048@xmath1083150 , pixel scale = 0435 per pixel ) .
sn 2009bb was observed with both johnson - kron - cousins @xmath111 and sloan @xmath112 filters . *
the swope telescope was also used to obtain near - infrared @xmath21 imaging with the retrocam camera ( hawaii-1 1024@xmath1081024 , pixel scale = 054 ) all optical images were reduced following standard procedures including bias , dark ( when appropriate ) , and flat - field corrections .
photometry of the supernova was computed relative to a local photometric sequence in the field of ngc 3278 .
the photometric sequence itself was calibrated to the standard johnson kron - cousins and sloan photometric systems using observations of @xcite and @xcite photometric standard stars obtained over the course of seven photometric nights .
our adopted @xmath15 and @xmath18 photometry of the local sequence is reported in table 7 and table 8 , respectively .
the listed values were computed via a weighted average of the measurements made for each of the seven calibration nights . given that the background of sn 2009bb was quite complex due to the presence of a bright region , it was necessary to apply template subtractions to all of the optical images .
three template images for each filter were acquired with the prompt telescopes between 2010 may 2530 , or @xmath24 420 days after @xmath2 maximum brightness . to estimate the sn residual flux in these templates , we rescaled the @xmath15 light curves of sn 1998bw in order to obtain the best match with sn 2009bb between + 30 and + 50 days .
the brightness decline of sn 1998bw is slower than that of sn 2009bb ( see figure [ fig2.1 ] ) , therefore a rescaling computed around maximum light should grossly overestimate the late - time flux of sn 2009bb . interpolating the rescaled light curves at the epoch on which the sn 2009bb templates were acquired
, we estimate the sn 2009bb brightness to be : @xmath2 @xmath24 24.6 mag , @xmath113 @xmath24 24.3 mag , @xmath114 @xmath24 23.3 mag and @xmath115 @xmath24 23.1 mag . assuming zero flux in the templates
, these values imply an underestimation of the sn 2009bb real flux in all the bands of less than 0.5% around maximum , and less than 2% on the last measurement published in this paper .
these values make us confident that the analysis reported here is not affected by systematic errors related to template subtraction . using the sn 2009bb spectra acquired on 2010 jan
. 9 ut and the standard bands of the johnson - kron - cousins and sloan photometric systems , we can estimate the previously mentioned systematic errors also for the @xmath116 prompt light curves .
as expected , also in this case the flux underestimation turns out to be negligible . taking into account that the decline rate of sn 1998bw is slower than sn
2009bb the previously reported sn 2009bb late phase magnitudes should be regarded as an upper limit of its real brightness .
nevertheless , possible late time flattening on the light curve can not be excluded a priori . in order to obtain a direct estimate of the sn 2009bb residual flux in the template images
, we examined the imacs @xmath117 images taken on 28 dec .
we measured the flux in an aperture centered in the sn position and of a radius equal to 3 times the sigma of the gaussian profile of the stellar objects present in the image . through these measurements
, we estimate upper limits of 20.9 mag , 20.5 mag and 20.1 mag for the @xmath117 filters respectively .
we note that most of the flux measured in the imacs images is due to the galaxy emission , not from the sn itself . to prove this , we computed synthetic magnitudes on the sn 2009bb nebular spectrum acquired on 9 jan .
2010 ut once this was cleaned from the region and stellar emissions .
we obtained : @xmath2=24.0 , @xmath113=23.9 , @xmath114=23.0 , @xmath115=23.1 , @xmath109=23.84 , @xmath118=23.4 , @xmath119=23.4 and @xmath120=23.8 mag . in the case of the swope telescope ,
the template images were acquired on 20 nov 2009 ut .
the sn flux in the @xmath20 and @xmath2 bands , was negligible , but this was not the case in the @xmath121 images . in order to reliably calibrate the magnitudes in the light curve tail
, we estimated the residual flux in the template images matching the prompt and swope light curves around maximum .
we next proceeded to compute @xmath122 photometry of sn 2009bb in the standard johnson and sloan systems using the s - correction technique @xcite , but following the prescription of @xcite .
since the @xmath20 band is not fully covered by any of our sn 2009bb spectra , the sn magnitudes were calibrated in the standard photometric system using the following equation : the near - infrared images were obtained using a standard jitter technique . to reduce the data cubes
, we made use of a pipeline developed by the csp ( for details see * ? ? ?
the pipeline applies to each image dark , flat - field , and detector linearity corrections .
these altered images are geometrically aligned and then stacked to obtain a master image from which photometry was computed .
near - infrared photometry of the sn was also obtained relative to a local sequence of stars calibrated against @xcite standard stars obtained during three photometric nights .
the @xmath21 photmetry of the local sequence stars are reported in table 9 . comparing our @xmath22 and @xmath23 magnitudes with these of the 2mass catalog
, we obtain a difference of 0.03 @xmath3 0.03 mag and 0.02 @xmath3 0.05 mag , respectively .
the near - infrared photometry instrumental magnitudes were computed through the template subtraction technique . as the photometry of the sn
is computed differentially with respect to the local sequence , we obtained final magnitudes using the following equations : the uvot images were reduced using the customized instrument pipeline that , in addition to preparing the frames for photometry , computes an astrometric solution for them .
aperture photometry was performed on the ra , dec position of sn 2009bb with the optimum aperture radius of 10 suggested by @xcite . to ensure that the aperture includes a constant amount of light , the stability of the psf size was checked in all the combined images .
the @xmath133 of the fwhm variation was 0.2 pixels for both the @xmath25 and @xmath26 uvot filters .
optical spectroscopy of sn 2009bb was obtained on thirteen epochs with the du pont ( + wfccd & b&c ) and magellan ( + imacs & ldss3 ) telescopes at lco , and the gemini south ( + gmos ) telescope .
spectra were reduced in a standard manner using iraf scripts based on tasks contained within the iraf noao.twodspec and gemini.gmos packages .
optimal extraction was obtained by weighting the signal according to the intensity profile along the slit .
sky subtraction was generally carried out by fitting a low order polynomial to both sides of the extracted sn spectrum , and wavelength solutions were determined from exposures of arc lamps .
the wavelength calibration was checked against bright night - sky emission lines .
flux calibration was performed by means of spectrophotometric standard stars @xcite and checked against the photometry .
when discrepancies occurred , the flux of the spectrum was scaled to match the broad - band photometry .
turatto m. , benetti s. & cappellaro e. , 2003 , in the proceedings to the eso / mpa / mpe workshop ( an eso astrophysics symposium ) from twilight to highlight : the physics of supernovae , eds .
b. leibundgut and w. hillebrandt ( springer - verlag ) , p. 200
band magnitude scale .
the red filled circles are the measurements considered in the fit of the fireball model ( solid line ) used to estimate the sn explosion date .
the first point indicates the upper - limit on the image taken immediately previous to the sn detection . ]
cccccccc [ tab1 ] 24/3/2009 & 2454914.5 & 18.10 @xmath3 0.06 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 24/3/2009 & 2454914.5 & @xmath0 & 17.21 @xmath3 0.06 & 17.09 @xmath3 0.03 & 16.75 @xmath3 0.08 & prompt5 + 24/3/2009 & 2454914.6 & 18.00 @xmath3 0.1 & @xmath0 & @xmath0 & @xmath0 & swope + 25/3/2009 & 2454915.6 & 17.49 @xmath3 5.27 & 16.82 @xmath3 0.1 & @xmath0 & @xmath0 & swope + 26/3/2009 & 2454916.5 & 17.33 @xmath3 0.04 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 26/3/2009 & 2454916.5 & @xmath0 & 16.65 @xmath3 0.04 & 16.36 @xmath3 0.03 & 15.95 @xmath3 0.05 & prompt5 + 27/3/2009 & 2454917.5 & 17.21 @xmath3 0.03 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 27/3/2009 & 2454917.5 & @xmath0 & 16.48 @xmath3 0.04 & 16.15 @xmath3 0.03 & 15.80 @xmath3 0.05 & prompt5 + 27/3/2009 & 2454917.7 & 17.17 @xmath3 0.04 & 16.40 @xmath3 0.1 & @xmath0 & @xmath0 & swope + 28/3/2009 & 2454918.5 & 17.13 @xmath3 0.04 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 28/3/2009 & 2454918.5 & @xmath0 & 16.34 @xmath3 0.03 & 16.05 @xmath3 0.03 & @xmath0 & prompt5 + 28/3/2009 & 2454918.6 & @xmath0 & @xmath0 & @xmath0 & 15.71 @xmath3 0.04 & prompt5 + 29/3/2009 & 2454919.7 & 17.08 @xmath3 0.04 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 29/3/2009 & 2454919.7 & @xmath0 & 16.21 @xmath3 0.03 & @xmath0 & 15.59 @xmath3 0.06 & prompt5 + 30/3/2009 & 2454920.5 & 17.07 @xmath3 0.04 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 30/3/2009 & 2454920.5 & @xmath0 & 16.21 @xmath3 0.04 & 15.90 @xmath3 0.03 & 15.50 @xmath3 0.05 & prompt5 + 30/3/2009 & 2454920.6 & 17.07 @xmath3 0.03 & 16.16 @xmath3 0.1 & @xmath0 & @xmath0 & swope + 31/3/2009 & 2454921.5 & @xmath0 & 16.10 @xmath3 0.03 & 15.85 @xmath3 0.03 & 15.49 @xmath3 0.05 & prompt5 + 1/4/2009 & 2454922.5 & 17.07 @xmath3 0.03 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 1/4/2009 & 2454922.5 & @xmath0 & 16.08 @xmath3 0.03 & 15.81 @xmath3 0.03 & 15.42 @xmath3 0.06 & prompt5 + 2/4/2009 & 2454923.6 & @xmath0 & @xmath0 & @xmath0 & 15.35 @xmath3 0.04 & prompt5 + 2/4/2009 & 2454923.7 & 17.16 @xmath3 0.04 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 2/4/2009 & 2454923.7 & @xmath0 & 16.07 @xmath3 0.03 & 15.79 @xmath3 0.03 & @xmath0 & prompt5 + 3/4/2009 & 2454924.5 & 17.25 @xmath3 0.04 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 3/4/2009 & 2454924.6 & @xmath0 & 16.08 @xmath3 0.03 & 15.78 @xmath3 0.03 & 15.35 @xmath3 0.06 & prompt5 + 6/4/2009 & 2454927.6 & 17.57 @xmath3 0.27 & 16.21 @xmath3 0.1 & @xmath0 & @xmath0 & swope + 7/4/2009 & 2454928.5 & 17.78 @xmath3 0.03 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 7/4/2009 & 2454928.5 & @xmath0 & 16.31 @xmath3 0.03 & 15.85 @xmath3 0.04 & 15.37 @xmath3 0.04 & prompt5 + 7/4/2009 & 2454928.6 & 17.75 @xmath3 0.07 & 16.31 @xmath3 0.1 & @xmath0 & @xmath0 & swope + 8/4/2009 & 2454929.5 & 17.84 @xmath3 0.04 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 8/4/2009 & 2454929.5 & @xmath0 & 16.41 @xmath3 0.04 & 15.90 @xmath3 0.03 & 15.37 @xmath3 0.05 & prompt5 + 9/4/2009 & 2454930.5 & 18.00 @xmath3 0.07 & 16.50 @xmath3 0.1 & @xmath0 & @xmath0 & swope + 9/4/2009 & 2454930.5 & 18.02 @xmath3 0.05 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 9/4/2009 & 2454930.5 & @xmath0 & 16.52 @xmath3 0.04 & 16.00 @xmath3 0.03 & 15.40 @xmath3 0.05 & prompt5 + 10/4/2009 & 2454931.5 & @xmath0 & 16.59 @xmath3 0.04 & 16.01 @xmath3 0.03 & 15.44 @xmath3 0.04 & prompt5 + 10/4/2009 & 2454931.6 & 18.13 @xmath3 0.05 & 16.56 @xmath3 0.1 & @xmath0 & @xmath0 & swope + 11/4/2009 & 2454932.5 & @xmath0 & 16.71 @xmath3 0.04 & 16.10 @xmath3 0.03 & 15.48 @xmath3 0.05 & prompt5 + 12/4/2009 & 2454933.5 & @xmath0 & 16.76 @xmath3 0.05 & 16.14 @xmath3 0.03 & @xmath0 & prompt5 + 12/4/2009 & 2454933.6 & 18.40 @xmath3 0.10 & 16.75 @xmath3 0.1 & @xmath0 & @xmath0 & swope + 13/4/2009 & 2454934.5 & 18.48 @xmath3 0.05 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 13/4/2009 & 2454934.5 & @xmath0 & 16.90 @xmath3 0.04 & 16.22 @xmath3 0.05 & 15.59 @xmath3 0.04 & prompt5 + 14/4/2009 & 2454935.5 & 18.59 @xmath3 0.06 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 14/4/2009 & 2454935.5 & @xmath0 & 16.95 @xmath3 0.05 & 16.33 @xmath3 0.03 & 15.67 @xmath3 0.05 & prompt5 + 15/4/2009 & 2454936.5 & 18.67 @xmath3 0.05 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 15/4/2009 & 2454936.5 & @xmath0 & 17.11 @xmath3 0.05 & 16.43 @xmath3 0.03 & 15.74 @xmath3 0.04 & prompt5 + 16/4/2009 & 2454937.6 & 18.81 @xmath3 0.09 & 17.15 @xmath3 0.1 & @xmath0 & @xmath0 & swope + 17/4/2009 & 2454938.5 & @xmath0 & 17.25 @xmath3 0.06 & 16.60 @xmath3 0.03 & 15.89 @xmath3 0.05 & prompt5 + 17/4/2009 & 2454938.6 & 18.92 @xmath3 0.14 & 17.22 @xmath3 0.1 & @xmath0 & @xmath0 & swope + 18/4/2009 & 2454939.5 & @xmath0 & 17.35 @xmath3 0.06 & 16.68 @xmath3 0.03 & 15.96 @xmath3 0.05 & prompt5 + 20/4/2009 & 2454941.6 & @xmath0 & 17.54 @xmath3 0.06 & 16.82 @xmath3 0.03 & 16.12 @xmath3 0.05 & prompt5 + 21/4/2009 & 2454942.5 & @xmath0 & 17.54 @xmath3 0.07 & 16.92 @xmath3 0.03 & 16.16 @xmath3 0.05 & prompt5 + 21/4/2009 & 2454942.6 & 19.16 @xmath3 0.12 & 17.51 @xmath3 0.1 & @xmath0 & @xmath0 & swope + 22/4/2009 & 2454943.5 & @xmath0 & 17.62 @xmath3 0.07 & 16.96 @xmath3 0.04 & 16.23 @xmath3 0.05 & prompt5 + 23/4/2009 & 2454944.6 & @xmath0 & 17.67 @xmath3 0.06 & @xmath0 & 16.38 @xmath3 0.05 & prompt5 + 23/4/2009 & 2454944.7 & 19.38 @xmath3 0.10 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 25/4/2009 & 2454946.5 & 19.38 @xmath3 0.09 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 25/4/2009 & 2454946.5 & @xmath0 & 17.74 @xmath3 0.07 & @xmath0 & @xmath0 & prompt5 + 25/4/2009 & 2454946.6 & @xmath0 & @xmath0 & 17.18 @xmath3 0.03 & 16.45 @xmath3 0.06 & prompt5 + 26/4/2009 & 2454947.5 & @xmath0 & 17.80 @xmath3 0.1 & @xmath0 & @xmath0 & swope + 26/4/2009 & 2454947.5 & @xmath0 & @xmath0 & @xmath0 & 16.47 @xmath3 0.05 & prompt5 + 26/4/2009 & 2454947.6 & 19.43 @xmath3 0.19 & @xmath0 & @xmath0 & @xmath0 & swope + 26/4/2009 & 2454947.6 & @xmath0 & 17.85 @xmath3 0.08 & 17.20 @xmath3 0.03 & @xmath0 & prompt5 + 27/4/2009 & 2454948.6 & @xmath0 & 17.88 @xmath3 0.07 & 17.24 @xmath3 0.03 & 16.54 @xmath3 0.06 & prompt5 + 29/4/2009 & 2454950.5 & @xmath0 & @xmath0 & 17.33 @xmath3 0.03 & @xmath0 & prompt5 + 29/4/2009 & 2454950.6 & @xmath0 & 17.95 @xmath3 0.11 & @xmath0 & 16.60 @xmath3 0.07 & prompt5 + 30/4/2009 & 2454951.5 & @xmath0 & 17.99 @xmath3 0.10 & 17.35 @xmath3 0.04 & 16.61 @xmath3 0.08 & prompt5 + 1/5/2009 & 2454952.5 & @xmath0 & 18.07 @xmath3 0.11 & 17.39 @xmath3 0.03 & 16.64 @xmath3 0.07 & prompt5 + 2/5/2009 & 2454953.5 & 19.48 @xmath3 0.04 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 3/5/2009 & 2454954.5 & @xmath0 & 18.15 @xmath3 0.10 & 17.50 @xmath3 0.04 & 16.73 @xmath3 0.07 & prompt5 + 4/5/2009 & 2454955.5 & @xmath0 & 18.16 @xmath3 0.11 & 17.50 @xmath3 0.03 & 16.81 @xmath3 0.08 & prompt5 + 5/5/2009 & 2454956.5 & 19.93 @xmath3 0.49 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 5/5/2009 & 2454956.5 & @xmath0 & 18.30 @xmath3 0.11 & 17.63 @xmath3 0.03 & 16.87 @xmath3 0.06 & prompt5 + 8/5/2009 & 2454959.5 & 19.70 @xmath3 0.20 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 8/5/2009 & 2454959.5 & @xmath0 & 18.21 @xmath3 0.09 & 17.70 @xmath3 0.03 & 16.95 @xmath3 0.08 & prompt5 + 9/5/2009 & 2454960.5 & 19.81 @xmath3 0.13 & 18.23 @xmath3 0.1 & @xmath0 & @xmath0 & swope + 9/5/2009 & 2454960.5 & 19.88 @xmath3 0.14 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 9/5/2009 & 2454960.5 & @xmath0 & 18.21 @xmath3 0.11 & 17.72 @xmath3 0.03 & @xmath0 & prompt5 + 9/5/2009 & 2454960.6 & @xmath0 & @xmath0 & @xmath0 & 16.86 @xmath3 0.08 & prompt5 + 10/5/2009 & 2454961.6 & 20.08 @xmath3 0.16 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 10/5/2009 & 2454961.6 & @xmath0 & 18.36 @xmath3 0.12 & 17.79 @xmath3 0.03 & 17.06 @xmath3 0.10 & prompt5 + 11/5/2009 & 2454962.5 & 19.86 @xmath3 0.27 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 11/5/2009 & 2454962.5 & @xmath0 & 18.31 @xmath3 0.14 & 17.71 @xmath3 0.04 & 16.94 @xmath3 0.07 & prompt5 + 12/5/2009 & 2454963.5 & 19.64 @xmath3 0.27 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 12/5/2009 & 2454963.5 & @xmath0 & 18.41 @xmath3 0.14 & 17.95 @xmath3 0.03 & 17.08 @xmath3 0.05 & prompt5 + 13/5/2009 & 2454964.5 & 19.91 @xmath3 0.29 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 13/5/2009 & 2454964.5 & @xmath0 & 18.75 @xmath3 0.17 & 17.90 @xmath3 0.03 & @xmath0 & prompt5 + 13/5/2009 & 2454964.6 & 19.96 @xmath3 0.1 & 18.34 @xmath3 0.1 & @xmath0 & @xmath0 & swope + 15/5/2009 & 2454966.5 &
20.03 @xmath3 0.1 & 18.48 @xmath3 0.1 & @xmath0 & @xmath0 & swope + 16/5/2009 & 2454967.5 & @xmath0 & 18.29 @xmath3 0.09 & 17.99 @xmath3 0.03 & 16.93 @xmath3 0.09 & prompt5 + 17/5/2009 & 2454968.5 & @xmath0 & 18.46 @xmath3 0.13 & 17.98 @xmath3 0.04 & 17.21 @xmath3 0.11 & prompt5 + 19/5/2009 & 2454970.5 & @xmath0 & 18.49 @xmath3 0.15 & 17.98 @xmath3 0.06 & 17.39 @xmath3 0.12 & prompt5 + 21/5/2009 & 2454972.5 & @xmath0 & 18.99 @xmath3 0.18 & 18.14 @xmath3 0.03 & 17.33 @xmath3 0.11 & prompt5 + 22/5/2009 & 2454973.5 & @xmath0 & 18.72 @xmath3 0.18 & 18.11 @xmath3 0.03 & 17.35 @xmath3 0.11 & prompt5 + 2/6/2009 & 2454984.5 & @xmath0 & 18.95 @xmath3 0.19 & 18.57 @xmath3 0.05 & @xmath0 & prompt5 + 2/6/2009 & 2454984.6 & @xmath0 & @xmath0 & @xmath0 & 17.47 @xmath3 0.17 & prompt5 + 11/6/2009 & 2454993.5 & @xmath0 & 19.14 @xmath3 0.17 & 18.58 @xmath3 0.03 & 18.11 @xmath3 0.20 & prompt5 + 12/6/2009 & 2454994.5 & @xmath0 & @xmath0 & 18.90 @xmath3 0.03 & @xmath0 & prompt5 + 12/6/2009 & 2454994.6 & @xmath0 & @xmath0 & @xmath0 & 18.00 @xmath3 0.23 & prompt5 + 13/6/2009 & 2454995.5 & @xmath0 & 19.59 @xmath3 0.29 & 18.84 @xmath3 0.04 & 18.00 @xmath3 0.17 & prompt5 + 25/7/2009 & 2455037.5 & @xmath0 & 19.78 @xmath3 0.78 & @xmath0 & @xmath0 & prompt5 + cccccccc [ tab1.2 ] 24/3/2009 & 2454914.5 & @xmath0 & @xmath0 & 17.33 @xmath3 0.03 & 17.24 @xmath3 0.05 & 17.12 @xmath3 0.08 & prompt5 + 24/3/2009 & 2454914.5 & @xmath0 & 17.58 @xmath3 0.06 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 24/3/2009 & 2454914.6 & 18.84 @xmath3 0.1 & 17.59 @xmath3 0.1 & 17.23 @xmath3 0.1 & 17.10 @xmath3 0.1 & @xmath0 & swope + 25/3/2009 & 2454915.6 & 18.43 @xmath3 0.1 & 17.21 @xmath3 0.03 & 16.82 @xmath3 0.03 & 16.71 @xmath3 0.05 & @xmath0 & swope + 26/3/2009 & 2454916.5 & @xmath0 & @xmath0 & 16.55 @xmath3 0.03 & 16.44 @xmath3 0.03 & 16.38 @xmath3 0.07 & prompt5 + 26/3/2009 & 2454916.5 & @xmath0 & 16.99 @xmath3 0.05 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 27/3/2009 & 2454917.5 & @xmath0 & @xmath0 & 16.35 @xmath3 0.03 & 16.25 @xmath3 0.03 & 16.07 @xmath3 0.06 & prompt5 + 27/3/2009 & 2454917.5 & @xmath0 & 16.84 @xmath3 0.04 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 27/3/2009 & 2454917.7 & 18.13 @xmath3 0.1 & 16.82 @xmath3 0.03 & 16.32 @xmath3 0.03 & 16.27 @xmath3 0.04 & @xmath0 & swope + 28/3/2009 & 2454918.5 & @xmath0 & @xmath0 & 16.20 @xmath3 0.03 & 16.18 @xmath3 0.03 & 15.95 @xmath3 0.07 & prompt5 + 28/3/2009 & 2454918.5 & @xmath0 & 16.72 @xmath3 0.04 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 29/3/2009 & 2454919.7 & @xmath0 & 16.66 @xmath3 0.04 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 30/3/2009 & 2454920.5 & @xmath0 & @xmath0 & 16.06 @xmath3 0.03 & 16.03 @xmath3 0.03 & 15.80 @xmath3 0.06 & prompt5 + 30/3/2009 & 2454920.5 & @xmath0 & 16.62 @xmath3 0.04 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 30/3/2009 & 2454920.6 & 18.11 @xmath3 0.1 & 16.63 @xmath3 0.03 & 16.07 @xmath3 0.03 & 16.00 @xmath3 0.04 & @xmath0 & swope + 31/3/2009 & 2454921.5 & @xmath0 & @xmath0 & 16.03 @xmath3 0.03 & 16.00 @xmath3 0.03 & 15.78 @xmath3 0.06 & prompt5 + 1/4/2009 & 2454922.5 & @xmath0 & @xmath0 & 16.02 @xmath3 0.03 & 15.95 @xmath3 0.03 & 15.69 @xmath3 0.06 & prompt5 + 1/4/2009 & 2454922.5 & @xmath0 & 16.65 @xmath3 0.04 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 2/4/2009 & 2454923.6 & @xmath0 & @xmath0 & 15.96 @xmath3 0.03 & 15.91 @xmath3 0.03 & 15.63 @xmath3 0.07 & prompt5 + 2/4/2009 & 2454923.7 & @xmath0 & 16.69 @xmath3 0.04 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 3/4/2009 & 2454924.5 & @xmath0 & @xmath0 & 15.98 @xmath3 0.03 & 15.91 @xmath3 0.03 & 15.60 @xmath3 0.06 & prompt5 + 3/4/2009 & 2454924.5 & @xmath0 & 16.70 @xmath3 0.04 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 6/4/2009 & 2454927.6 & 18.98 @xmath3 0.1 & 16.97 @xmath3 0.03 & 16.02 @xmath3 0.04 & 15.90 @xmath3 0.03 & @xmath0 & swope + 7/4/2009 & 2454928.5 & 19.27 @xmath3 0.1 & 17.09 @xmath3 0.03 & 16.05 @xmath3 0.03 & 15.93 @xmath3 0.03 & @xmath0 & swope + 7/4/2009 & 2454928.5 & @xmath0 & @xmath0 & 16.07 @xmath3 0.03 & 15.93 @xmath3 0.03 & 15.60 @xmath3 0.06 & prompt5 + 7/4/2009 & 2454928.5 & @xmath0 & 17.05 @xmath3 0.04 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 8/4/2009 & 2454929.5 & @xmath0 & @xmath0 & 16.14 @xmath3 0.03 & 15.96 @xmath3 0.03 & 15.70 @xmath3 0.06 & prompt5 + 8/4/2009 & 2454929.5 & @xmath0 & 17.14 @xmath3 0.05 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 9/4/2009 & 2454930.5 & 19.62 @xmath3 0.1 & 17.30 @xmath3 0.03 & 16.16 @xmath3 0.04 & 16.00 @xmath3 0.04 & @xmath0 & swope + 9/4/2009 & 2454930.5 & @xmath0 & @xmath0 & 16.15 @xmath3 0.03 & 16.00 @xmath3 0.04 & 15.64 @xmath3 0.07 & prompt5 + 9/4/2009 & 2454930.5 & @xmath0 & 17.30 @xmath3 0.05 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 10/4/2009 & 2454931.5 & 19.71 @xmath3 0.1 & 17.42 @xmath3 0.03 & 16.22 @xmath3 0.03 & 16.05 @xmath3 0.03 & @xmath0 & swope + 10/4/2009 & 2454931.5 & @xmath0 & @xmath0 & 16.22 @xmath3 0.03 & 16.05 @xmath3 0.03 & 15.69 @xmath3 0.06 & prompt5 + 10/4/2009 & 2454931.5 & @xmath0 & 17.38 @xmath3 0.05 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 11/4/2009 & 2454932.5 & @xmath0 & @xmath0 & 16.34 @xmath3 0.03 & 16.09 @xmath3 0.03 & 15.73 @xmath3 0.06 & prompt5 + 11/4/2009 & 2454932.5 & @xmath0 & 17.52 @xmath3 0.05 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 12/4/2009 & 2454933.5 & 20.03 @xmath3 0.1 & 17.67 @xmath3 0.03 & 16.37 @xmath3 0.03 & 16.15 @xmath3 0.04 & @xmath0 & swope + 12/4/2009 & 2454933.5 & @xmath0 & @xmath0 & 16.40 @xmath3 0.03 & @xmath0 & 15.77 @xmath3 0.06 & prompt5 + 12/4/2009 & 2454933.5 & @xmath0 & 17.62 @xmath3 0.05 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 13/4/2009 & 2454934.5 & @xmath0 & @xmath0 & 16.48 @xmath3 0.03 & 16.25 @xmath3 0.03 & 15.85 @xmath3 0.06 & prompt5 + 13/4/2009 & 2454934.5 & @xmath0 & 17.72 @xmath3 0.07 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 14/4/2009 & 2454935.5 & @xmath0 & @xmath0 & 16.55 @xmath3 0.03 & 16.30 @xmath3 0.04 & 15.90 @xmath3 0.06 & prompt5 + 14/4/2009 & 2454935.5 & @xmath0 & 17.92 @xmath3 0.07 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 15/4/2009 & 2454936.5 & @xmath0 & @xmath0 & 16.69 @xmath3 0.03 & 16.41 @xmath3 0.04 & 15.96 @xmath3 0.06 & prompt5 + 15/4/2009 & 2454936.5 & @xmath0 & 18.07 @xmath3 0.08 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 16/4/2009 & 2454937.6 & 20.51 @xmath3 0.1 & 18.15 @xmath3 0.03 & 16.75 @xmath3 0.03 & 16.49 @xmath3 0.04 & @xmath0 & swope + 17/4/2009 & 2454938.5 & @xmath0 & @xmath0 & 16.84 @xmath3 0.03 & 16.61 @xmath3 0.04 & 16.10 @xmath3 0.06 & prompt5 + 17/4/2009 & 2454938.6 & 20.48 @xmath3 0.1 & 18.25 @xmath3 0.03 & 16.86 @xmath3 0.03 & 16.58 @xmath3 0.03 & @xmath0 & swope + 18/4/2009 & 2454939.5 & @xmath0 & @xmath0 & 16.96 @xmath3 0.03 & 16.63 @xmath3 0.04 & 16.16 @xmath3 0.06 & prompt5 + 18/4/2009 & 2454939.5 & @xmath0 & 18.30 @xmath3 0.10 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 20/4/2009 & 2454941.6 & @xmath0 & @xmath0 & 17.10 @xmath3 0.03 & 16.81 @xmath3 0.04 & 16.31 @xmath3 0.07 & prompt5 + 20/4/2009 & 2454941.6 & @xmath0 & 18.55 @xmath3 0.13 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 21/4/2009 & 2454942.5 & @xmath0 & @xmath0 & 17.23 @xmath3 0.03 & 16.83 @xmath3 0.04 & 16.36 @xmath3 0.07 & prompt5 + 21/4/2009 & 2454942.5 & @xmath0 & 18.62 @xmath3 0.13 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 21/4/2009 & 2454942.6 & 20.65 @xmath3 0.1 & 18.55 @xmath3 0.03 & 17.16 @xmath3 0.03 & 16.89 @xmath3 0.06 & @xmath0 & swope + 22/4/2009 & 2454943.5 & @xmath0 & @xmath0 & 17.27 @xmath3 0.03 & 16.94 @xmath3 0.04 & 16.41 @xmath3 0.07 & prompt5 + 22/4/2009 & 2454943.5 & @xmath0 & 18.67 @xmath3 0.13 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 23/4/2009 & 2454944.6 & @xmath0 & @xmath0 & 17.38 @xmath3 0.04 & @xmath0 & 16.52 @xmath3 0.07 & prompt5 + 23/4/2009 & 2454944.6 & @xmath0 & 18.80 @xmath3 0.15 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 25/4/2009 & 2454946.5 & @xmath0 & 18.83 @xmath3 0.16 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 25/4/2009 & 2454946.6 & @xmath0 & @xmath0 & 17.47 @xmath3 0.03 & 17.08 @xmath3 0.04 & 16.53 @xmath3 0.07 & prompt5 + 26/4/2009 & 2454947.5 & 21.05 @xmath3 0.1 & 18.98 @xmath3 0.06 & 17.54 @xmath3 0.03 & 17.24 @xmath3 0.05 & @xmath0 & swope + 26/4/2009 & 2454947.5 & @xmath0 & @xmath0 & 17.48 @xmath3 0.03 & 17.21 @xmath3 0.05 & 16.58 @xmath3 0.07 & prompt5 + 26/4/2009 & 2454947.5 & @xmath0 & 18.98 @xmath3 0.14 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 27/4/2009 & 2454948.5 & @xmath0 & 18.98 @xmath3 0.18 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 27/4/2009 & 2454948.6 & @xmath0 & @xmath0 & 17.56 @xmath3 0.04 & 17.24 @xmath3 0.05 & 16.68 @xmath3 0.07 & prompt5 + 28/4/2009 & 2454949.5 & @xmath0 & 19.06 @xmath3 0.19 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 29/4/2009 & 2454950.5 & @xmath0 & @xmath0 & @xmath0 & @xmath0 & 16.71 @xmath3 0.08 & prompt5 + 29/4/2009 & 2454950.6 & @xmath0 & @xmath0 & 17.69 @xmath3 0.04 & 17.38 @xmath3 0.05 & @xmath0 & prompt5 + 29/4/2009 & 2454950.6 & @xmath0 & 19.12 @xmath3 0.20 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 30/4/2009 & 2454951.5 & @xmath0 & @xmath0 & 17.75 @xmath3 0.04 & 17.43 @xmath3 0.05 & 16.79 @xmath3 0.08 & prompt5 + 30/4/2009 & 2454951.5 & @xmath0 & 19.29 @xmath3 0.20 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 1/5/2009 & 2454952.5 & @xmath0 & @xmath0 & 17.83 @xmath3 0.04 & 17.50 @xmath3 0.05 & 16.79 @xmath3 0.08 & prompt5 + 1/5/2009 & 2454952.5 & @xmath0 & 19.32 @xmath3 0.23 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 2/5/2009 & 2454953.5 & @xmath0 & 19.26 @xmath3 0.18 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 3/5/2009 & 2454954.5 & @xmath0 & @xmath0 & 17.88 @xmath3 0.03 & 17.61 @xmath3 0.06 & 16.90 @xmath3 0.08 & prompt5 + 3/5/2009 & 2454954.5 & @xmath0 & 19.37 @xmath3 0.17 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 4/5/2009 & 2454955.5 & @xmath0 & @xmath0 & 17.93 @xmath3 0.04 & 17.57 @xmath3 0.05 & 16.85 @xmath3 0.09 & prompt5 + 5/5/2009 & 2454956.5 & @xmath0 & @xmath0 & 17.97 @xmath3 0.06 & @xmath0 & 16.96 @xmath3 1.84 & prompt5 + 8/5/2009 & 2454959.5 & @xmath0 & @xmath0 & 18.17 @xmath3 0.05 & 17.70 @xmath3 0.06 & 17.08 @xmath3 0.10 & prompt5 + 8/5/2009 & 2454959.5 & @xmath0 & 19.46 @xmath3 0.20 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 9/5/2009 & 2454960.5 & 21.62 @xmath3 0.1 & 19.49 @xmath3 0.03 & 18.20 @xmath3 0.07 & 17.80 @xmath3 0.07 & @xmath0 & swope + 9/5/2009 & 2454960.5 & @xmath0 & @xmath0 & 18.18 @xmath3 0.06 & 17.69 @xmath3 0.05 & @xmath0 & prompt5 + 9/5/2009 & 2454960.5 & @xmath0 & 19.43 @xmath3 0.46 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 9/5/2009 & 2454960.6 & @xmath0 & @xmath0 & @xmath0 & @xmath0 & 17.04 @xmath3 0.08 & prompt5 + 10/5/2009 & 2454961.5 & @xmath0 & @xmath0 & 18.12 @xmath3 0.04 & @xmath0 & @xmath0 & prompt5 + 10/5/2009 & 2454961.6 & @xmath0 & @xmath0 & @xmath0 & 17.84 @xmath3 0.07 & 17.11 @xmath3 0.07 & prompt5 + 10/5/2009 & 2454961.6 & @xmath0 & 19.53 @xmath3 0.28 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 11/5/2009 & 2454962.5 & @xmath0 & @xmath0 & @xmath0 & 17.86 @xmath3 0.06 & 17.07 @xmath3 0.09 & prompt5 + 12/5/2009 & 2454963.5 & @xmath0 & @xmath0 & 18.19 @xmath3 0.04 & 17.89 @xmath3 0.05 & 17.17 @xmath3 0.11 & prompt5 + 13/5/2009 & 2454964.5 & @xmath0 & 19.51 @xmath3 0.33 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 13/5/2009 & 2454964.6 & 21.61 @xmath3 0.1 & 19.59 @xmath3 0.04 & 18.30 @xmath3 0.05 & 17.96 @xmath3 0.10 & @xmath0 & swope + 15/5/2009 & 2454966.5 & @xmath0 & 19.65 @xmath3 0.15 & 18.43 @xmath3 0.19 & 18.00 @xmath3 0.07 & @xmath0 & swope + 17/5/2009 & 2454968.5 & @xmath0 & 19.88 @xmath3 0.70 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 21/5/2009 & 2454972.5 & @xmath0 & @xmath0 & 18.60 @xmath3 0.06 & 18.22 @xmath3 0.08 & 17.48 @xmath3 0.10 & prompt5 + 22/5/2009 & 2454973.6 & @xmath0 & @xmath0 & 18.51 @xmath3 0.05 & 18.29 @xmath3 0.09 & 17.53 @xmath3 0.10 & prompt5 + 22/5/2009 & 2454973.6 & @xmath0 & 19.95 @xmath3 0.48 & @xmath0 & @xmath0 & @xmath0 & prompt3 + 23/5/2009 & 2454974.5 & @xmath0 & @xmath0 & 18.69 @xmath3 0.07 & 18.25 @xmath3 0.07 & 17.63 @xmath3 0.13 & prompt5 + cccccccc [ tab7.1 ] 1 & 10:31:45.355 & -40:00:42.73 & 15.26 @xmath3 0.01 & 14.54 @xmath3 0.04 & 14.13 @xmath3 0.02 & 13.76 @xmath3 0.01 + 2 & 10:31:44.107 & -39:55:25.68 & 15.44 @xmath3 0.02 & 14.47 @xmath3 0.04 & 13.93 @xmath3 0.03 & 13.41 @xmath3 0.02 + 3 & 10:31:42.905 & -39:57:51.19 & 13.83 @xmath3 0.03 & 13.18 @xmath3 0.03 & 12.80 @xmath3 0.02 & 12.45 @xmath3 0.01 + 4 & 10:31:41.494 & -39:57:10.26 & 15.50 @xmath3 0.02 & 14.53 @xmath3 0.04 & 13.99 @xmath3 0.02 & 13.49 @xmath3 0.02 + 5 & 10:31:36.185 & -39:54:45.04 & 14.98 @xmath3 0.01 & 14.36 @xmath3 0.03 & 13.99 @xmath3 0.02 & 13.63 @xmath3 0.02 + 6 & 10:31:24.828 & -39:58:30.29 & 14.63 @xmath3 0.01 & 14.12 @xmath3 0.03 & 13.77 @xmath3 0.01 & 13.42 @xmath3 0.01 + 7 & 10:31:24.547 & -40:01:44.72 & 15.00 @xmath3 0.01 & 14.48 @xmath3 0.04 & 14.14 @xmath3 0.02 & 13.79 @xmath3 0.01 + 8 & 10:31:50.942 & -39:57:50.11 & 15.90 @xmath3 0.01 & 15.04 @xmath3 0.01 & 14.51 @xmath3 0.03 & 14.02 @xmath3 0.01 + 9 & 10:31:20.933 & -39:59:52.55 & 15.83 @xmath3 0.02 & 15.02 @xmath3 0.01 & 14.57 @xmath3 0.03 & 14.15 @xmath3 0.01 + 10 & 10:31:48.010 & -39:54:23.65 & 15.52 @xmath3 0.01 & 14.96 @xmath3 0.01 & 14.64 @xmath3 0.02 & 14.29 @xmath3 0.02 + 11 & 10:31:33.866 & -40:00:09.68 & 15.86 @xmath3 0.02 & 15.09 @xmath3 0.01 & 14.69 @xmath3 0.02 & 14.29 @xmath3 0.01 + 12 & 10:31:26.273 & -39:56:48.23 & 15.75 @xmath3 0.01 & 15.10 @xmath3 0.01 & 14.75 @xmath3 0.01 & 14.38 @xmath3 0.02 + 13 & 10:31:45.754 & -39:56:32.17 & 15.84 @xmath3 0.01 & 15.21 @xmath3 0.01 & 14.86 @xmath3 0.02 & 14.51 @xmath3 0.03 + 14 & 10:31:51.468 & -39:54:42.59 & 16.25 @xmath3 0.01 & 15.52 @xmath3 0.01 & 15.10 @xmath3 0.03 & 14.68 @xmath3 0.03 + 15 & 10:31:51.929 & -39:54:27.94 & 16.36 @xmath3 0.01 & 15.56 @xmath3 0.01 & 15.11 @xmath3 0.02 & 14.63 @xmath3 0.04 + 16 & 10:31:24.012 & -39:59:29.94 & 16.41 @xmath3 0.02 & 15.62 @xmath3 0.01 & 15.20 @xmath3 0.03 & 14.78 @xmath3 0.02 + 17 & 10:31:14.064 & -39:58:45.55 & 16.45 @xmath3 0.03 & 15.64 @xmath3 0.01 & 15.19 @xmath3 0.03 & 14.77 @xmath3 0.01 + 18 & 10:31:16.582 & -39:56:39.70 & 16.13 @xmath3 0.04 & 15.62 @xmath3 0.04 & 15.37 @xmath3 0.05 & 15.06 @xmath3 0.06 + 19 & 10:31:18.864 & -39:57:51.30 & 16.55 @xmath3 0.01 & 15.76 @xmath3 0.01 & 15.33 @xmath3 0.02 & 14.91 @xmath3 0.01 + 20 & 10:31:23.846 & -39:54:50.40 & 16.49 @xmath3 0.02 & 15.80 @xmath3 0.01 & 15.40 @xmath3 0.02 & 15.01 @xmath3 0.04 + 21 & 10:31:24.442 & -39:55:08.33 & 16.79 @xmath3 0.03 & 15.96 @xmath3 0.01 & 15.49 @xmath3 0.01 & 15.07 @xmath3 0.03 + 22 & 10:31:46.673 & -39:58:40.19 & 16.98 @xmath3 0.01 & 16.28 @xmath3 0.01 & 15.89 @xmath3 0.02 & 15.50 @xmath3 0.02 + 23 & 10:31:31.198 & -39:53:32.46 & 17.52 @xmath3 0.02 & 16.62 @xmath3 0.01 & 16.06 @xmath3 0.05 & 15.56 @xmath3 0.02 + 24 & 10:31:49.822 & -40:00:16.78 & 18.60 @xmath3 0.01 & 17.08 @xmath3 0.03 & 16.12 @xmath3 0.06 & 15.04 @xmath3 0.02 + 25 & 10:31:42.703 & -39:58:20.78 & 17.56 @xmath3 0.01 & 16.90 @xmath3 0.01 & 16.50 @xmath3 0.03 & 16.06 @xmath3 0.03 + 26 & 10:31:48.434 & -40:01:29.93 & 17.84 @xmath3 0.02 & 17.14 @xmath3 0.01 & 16.72 @xmath3 0.05 & 16.32 @xmath3 0.05 + 27 & 10:31:54.881 & -39:56:16.58 & 19.25 @xmath3 0.02 & 17.78 @xmath3 0.01 & 16.77 @xmath3 0.14 & 15.98 @xmath3 0.10 + 28 & 10:31:33.463 & -39:58:54.05 & 18.11 @xmath3 0.02 & 17.45 @xmath3 0.01 & 17.10 @xmath3 0.04 & 16.71 @xmath3 0.02 + 29 & 10:31:37.385 & -39:57:52.78 & 18.83 @xmath3 0.02 & 17.86 @xmath3 0.01 & 17.35 @xmath3 0.09 & 16.82 @xmath3 0.04 + 30 & 10:31:32.995 & -39:58:26.69 & 19.39 @xmath3 0.02 & 18.22 @xmath3 0.01 & 17.52 @xmath3 0.04 & 16.76 @xmath3 0.21 + 31 & 10:31:31.990 & -39:58:28.85 & 18.86 @xmath3 0.01 & 18.11 @xmath3 0.01 & 17.64 @xmath3 0.06 & 17.21 @xmath3 0.06 + ccccccccc [ tab7.2 ] 1 & 10:31:45.355 & -40:00:42.73 & 16.37 @xmath3 0.06 & 14.83 @xmath3 0.01 & 14.31 @xmath3 0.02 & 14.14 @xmath3 0.03 & 14.09 @xmath3 0.02 + 2 & 10:31:44.107 & -39:55:25.68 & 17.04 @xmath3 0.07 & @xmath0 & 14.13 @xmath3 0.02 & 13.82 @xmath3 0.03 & 13.71 @xmath3 0.01 + 3 & 10:31:42.905 & -39:57:51.19 & @xmath0 & @xmath0 & 12.96 @xmath3 0.02 & 12.81 @xmath3 0.03 & 12.81 @xmath3 0.02 + 4 & 10:31:41.494 & -39:57:10.26 & 17.20 @xmath3 0.06 & @xmath0 & 14.19 @xmath3 0.02 & 13.91 @xmath3 0.03 & 13.80 @xmath3 0.04 + 5 & 10:31:36.185 & -39:54:45.04 & @xmath0 & @xmath0 & 14.06 @xmath3 0.02 & 13.91 @xmath3 0.03 & 13.89 @xmath3 0.04 + 6 & 10:31:24.828 & -39:58:30.29 & @xmath0 & @xmath0 & 13.91 @xmath3 0.02 & 13.76 @xmath3 0.03 & 13.75 @xmath3 0.01 + 7 & 10:31:24.547 & -40:01:44.72 & 15.82 @xmath3 0.02 & 14.68 @xmath3 0.01 & 14.31 @xmath3 0.02 & 14.16 @xmath3 0.03 & 14.14 @xmath3 0.03 + 8 & 10:31:50.942 & -39:57:50.11 & 17.25 @xmath3 0.05 & 15.41 @xmath3 0.01 & 14.71 @xmath3 0.01 & 14.43 @xmath3 0.01 & 14.32 @xmath3 0.05 + 9 & 10:31:20.933 & -39:59:52.55 & 17.15 @xmath3 0.07 & 15.37 @xmath3 0.01 & 14.76 @xmath3 0.01 & 14.56 @xmath3 0.01 & 14.52 @xmath3 0.02 + 10 & 10:31:48.010 & -39:54:23.65 & 16.38 @xmath3 0.04 & 15.19 @xmath3 0.01 & 14.79 @xmath3 0.01 & 14.65 @xmath3 0.01 & 14.66 @xmath3 0.06 + 11 & 10:31:33.866 & -40:00:09.68 & 17.09 @xmath3 0.06 & 15.41 @xmath3 0.01 & 14.86 @xmath3 0.01 & 14.67 @xmath3 0.01 & 14.67 @xmath3 0.06 + 12 & 10:31:26.273 & -39:56:48.23 & 16.69 @xmath3 0.04 & 15.37 @xmath3 0.01 & 14.91 @xmath3 0.01 & 14.76 @xmath3 0.01 & 14.73 @xmath3 0.04 + 13 & 10:31:45.754 & -39:56:32.17 & 16.83 @xmath3 0.05 & 15.47 @xmath3 0.01 & 15.03 @xmath3 0.01 & 14.88 @xmath3 0.01 & 14.96 @xmath3 0.07 + 14 & 10:31:51.468 & -39:54:42.59 & 17.24 @xmath3 0.04 & 15.83 @xmath3 0.01 & 15.28 @xmath3 0.01 & 15.06 @xmath3 0.01 & 15.03 @xmath3 0.04 + 15 & 10:31:51.929 & -39:54:27.94 & 17.55 @xmath3 0.05 & 15.90 @xmath3 0.01 & 15.29 @xmath3 0.01 & 15.03 @xmath3 0.01 & 14.95 @xmath3 0.02 + 16 & 10:31:24.012 & -39:59:29.94 & 17.70 @xmath3 0.06 & 15.96 @xmath3 0.01 & 15.34 @xmath3 0.01 & 15.14 @xmath3 0.01 & 14.88 @xmath3 0.04 + 17 & 10:31:14.064 & -39:58:45.55 & 17.76 @xmath3 0.05 & 15.99 @xmath3 0.01 & 15.39 @xmath3 0.01 & 15.18 @xmath3 0.01 & 15.11 @xmath3 0.03 + 18 & 10:31:16.582 & -39:56:39.70 & 16.94 @xmath3 0.11 & 15.86 @xmath3 0.11 & 15.45 @xmath3 0.05 & 15.32 @xmath3 0.06 & 15.37 @xmath3 0.04 + 19 & 10:31:18.864 & -39:57:51.30 & 17.60 @xmath3 0.03 & 16.08 @xmath3 0.01 & 15.51 @xmath3 0.01 & 15.31 @xmath3 0.01 & 15.24 @xmath3 0.04 + 20 & 10:31:23.846 & -39:54:50.40 & 17.52 @xmath3 0.04 & 16.09 @xmath3 0.01 & 15.60 @xmath3 0.01 & 15.42 @xmath3 0.01 & 15.43 @xmath3 0.08 + 21 & 10:31:24.442 & -39:55:08.33 & 18.08 @xmath3 0.06 & 16.32 @xmath3 0.01 & 15.70 @xmath3 0.01 & 15.49 @xmath3 0.01 & 15.52 @xmath3 0.06 + 22 & 10:31:46.673 & -39:58:40.19 & 18.06 @xmath3 0.05 & 16.57 @xmath3 0.01 & 16.07 @xmath3 0.01 & 15.90 @xmath3 0.01 & 15.89 @xmath3 0.08 + 23 & 10:31:31.198 & -39:53:32.46 & 18.99 @xmath3 0.13 & 17.02 @xmath3 0.01 & 16.29 @xmath3 0.01 & 16.00 @xmath3 0.01 & 15.96 @xmath3 0.08 + 24 & 10:31:49.822 & -40:00:16.78 & 20.87 @xmath3 0.17 & 17.81 @xmath3 0.01 & 16.44 @xmath3 0.01 & 15.61 @xmath3 0.01 & 15.29 @xmath3 0.06 + 25 & 10:31:42.703 & -39:58:20.78 & 18.47 @xmath3 0.06 & 17.18 @xmath3 0.01 & 16.70 @xmath3 0.01 & 16.50 @xmath3 0.01 & 16.75 @xmath3 0.45 + 26 & 10:31:48.434 & -40:01:29.93 & 18.87 @xmath3 0.03 & 17.42 @xmath3 0.01 & 16.90 @xmath3 0.01 & 16.72 @xmath3 0.01 & 16.73 @xmath3 0.22 + 27 & 10:31:54.881 & -39:56:16.58 & @xmath0 & 18.50 @xmath3 0.01 & 17.14 @xmath3 0.01 & 16.42 @xmath3 0.01 & 16.04 @xmath3 0.08 + 28 & 10:31:33.463 & -39:58:54.05 & 19.05 @xmath3 0.05 & 17.73 @xmath3 0.01 & 17.25 @xmath3 0.01 & 17.07 @xmath3 0.01 & 17.54 @xmath3 0.92 + 29 & 10:31:37.385 & -39:57:52.78 & 20.41 @xmath3 0.08 & 18.31 @xmath3 0.01 & 17.46 @xmath3 0.01 & 17.16 @xmath3 0.01 & 17.91 @xmath3 0.46 + 30 & 10:31:32.995 & -39:58:26.69 & @xmath0 & 18.76 @xmath3 0.01 & 17.66 @xmath3 0.01 & 17.26 @xmath3 0.01 & 16.98 @xmath3 0.45 + 31 & 10:31:31.990 & -39:58:28.85 & 19.97 @xmath3 0.13 & 18.44 @xmath3 0.01 & 17.85 @xmath3 0.01 & 17.63 @xmath3 0.01 & 17.19 @xmath3 0.43 + cccccccc [ tab7.3 ] 1 & 10:31:45.355 & -40:00:42.73 & 12.23 @xmath3 0.01 & 11.99 @xmath3 0.01 & 11.70 @xmath3 0.01 + 2 & 10:31:44.107 & -39:55:25.68 & 12.78 @xmath3 0.01 & 12.57 @xmath3 0.01 & 12.34 @xmath3 0.02 + 3 & 10:31:42.905 & -39:57:51.19 & 13.03 @xmath3 0.01 & 12.69 @xmath3 0.01 & 12.19 @xmath3 0.01 + 4 & 10:31:41.494 & -39:57:10.26 & 13.11 @xmath3 0.01 & 12.76 @xmath3 0.01 & 12.29 @xmath3 0.02 + 5 & 10:31:36.185 & -39:54:45.04 & 13.21 @xmath3 0.01 & 12.97 @xmath3 0.01 & 12.68 @xmath3 0.01 + 6 & 10:31:24.828 & -39:58:30.29 & 13.22 @xmath3 0.01 & 12.85 @xmath3 0.01 & 12.37 @xmath3 0.02 + 7 & 10:31:24.547 & -40:01:44.72 & 13.55 @xmath3 0.01 & 13.27 @xmath3 0.01 & 12.92 @xmath3 0.01 + 8 & 10:31:50.942 & -39:57:50.11 & 13.41 @xmath3 0.01 & 13.16 @xmath3 0.01 & 12.87 @xmath3 0.03 + 9 & 10:31:20.933 & -39:59:52.55 & 13.69 @xmath3 0.01 & 13.32 @xmath3 0.01 & 12.75 @xmath3 0.01 + 10 & 10:31:48.010 & -39:54:23.65 & 13.59 @xmath3 0.01 & 13.29 @xmath3 0.01 & 12.86 @xmath3 0.01 + 11 & 10:31:33.866 & -40:00:09.68 & 13.73 @xmath3 0.01 & 13.45 @xmath3 0.01 & 13.14 @xmath3 0.01 + 12 & 10:31:26.273 & -39:56:48.23 & 13.71 @xmath3 0.01 & 13.35 @xmath3 0.01 & 12.84 @xmath3 0.01 + 13 & 10:31:51.468 & -39:54:42.59 & 13.83 @xmath3 0.01 & 13.42 @xmath3 0.01 & 12.88 @xmath3 0.01 + 14 & 10:31:51.929 & -39:54:27.94 & 13.97 @xmath3 0.01 & 13.69 @xmath3 0.01 & 13.39 @xmath3 0.02 + 15 & 10:31:24.012 & -39:59:29.94 & 14.05 @xmath3 0.01 & 13.75 @xmath3 0.01 & 13.42 @xmath3 0.01 + 16 & 10:31:14.064 & -39:58:45.55 & 14.15 @xmath3 0.01 & 13.90 @xmath3 0.01 & 13.59 @xmath3 0.01 + 17 & 10:31:16.582 & -39:56:39.70 & 14.28 @xmath3 0.01 & 14.00 @xmath3 0.01 & 13.71 @xmath3 0.01 + 18 & 10:31:18.864 & -39:57:51.30 & 14.21 @xmath3 0.01 & 13.80 @xmath3 0.01 & 13.28 @xmath3 0.01 + 19 & 10:31:23.846 & -39:54:50.40 & 14.30 @xmath3 0.01 & 14.05 @xmath3 0.01 & 13.76 @xmath3 0.01 + 20 & 10:31:24.442 & -39:55:08.33 & 14.24 @xmath3 0.01 & 14.00 @xmath3 0.01 & 13.73 @xmath3 0.02 + 21 & 10:31:46.673 & -39:58:40.19 & 14.29 @xmath3 0.01 & 13.97 @xmath3 0.01 & 13.58 @xmath3 0.02 + 22 & 10:31:31.198 & -39:53:32.46 & 14.39 @xmath3 0.01 & 14.10 @xmath3 0.01 & 13.71 @xmath3 0.02 + 23 & 10:31:49.822 & -40:00:16.78 & 14.48 @xmath3 0.01 & 14.02 @xmath3 0.01 & 13.35 @xmath3 0.01 + 24 & 10:31:42.703 & -39:58:20.78 & 14.54 @xmath3 0.01 & 14.24 @xmath3 0.01 & 13.90 @xmath3 0.01 + 25 & 10:31:48.434 & -40:01:29.93 & 14.77 @xmath3 0.01 & 14.49 @xmath3 0.01 & 14.15 @xmath3 0.01 + 26 & 10:31:54.881 & -39:56:16.58 & 14.74 @xmath3 0.01 & 14.23 @xmath3 0.01 & 13.67 @xmath3 0.01 + 27 & 10:31:33.463 & -39:58:54.05 & 15.09 @xmath3 0.01 & 14.82 @xmath3 0.01 & 14.48 @xmath3 0.02 + 28 & 10:31:37.385 & -39:57:52.78 & 15.25 @xmath3 0.01 & 14.95 @xmath3 0.01 & 14.71 @xmath3 0.01 + 30 & 10:31:32.995 & -39:58:26.69 & 15.72 @xmath3 0.01 & 15.24 @xmath3 0.01 & 14.70 @xmath3 0.01 + 29 & 10:31:31.990 & -39:58:28.85 & 15.63 @xmath3 0.01 & 15.12 @xmath3 0.01 & 14.57 @xmath3 0.02 + 31 & 10:31:21.082 & -39:58:30.44 & 15.84 @xmath3 0.01 & 15.56 @xmath3 0.01 & 15.30 @xmath3 0.03 + | ultraviolet , optical , and near - infrared photometry and optical spectroscopy of the broad - lined type ic supernova ( sn ) 2009bb are presented , following the flux evolution from @xmath010 to @xmath1285 days past @xmath2-band maximum .
thanks to the very early discovery , it is possible to place tight constraints on the sn explosion epoch .
the expansion velocities measured from near maximum spectra are found to be only slightly smaller than those measured from spectra of the prototype broad - lined sn 1998bw associated with grb 980425 .
fitting an analytical model to the pseudo - bolometric light curve of sn 2009bb suggests that [email protected] m@xmath4 of material was ejected with 0.22 @xmath30.06 m@xmath4 of it being @xmath5ni .
the resulting kinetic energy is [email protected]@xmath6 erg .
this , together with an absolute peak magnitude of m@xmath7 @xmath8 @[email protected] , places sn 2009bb on the energetic and luminous end of the broad - lined type ic ( sn ic ) sequence .
detection of helium in the early time optical spectra accompanied with strong radio emission , and high metallicity of its environment makes sn 2009bb a peculiar object .
similar to the case for grbs , we find that the bulk explosion parameters of sn 2009bb can not account for the copious energy coupled to relativistic ejecta , and conclude that another energy reservoir ( a central engine ) is required to power the radio emission . nevertheless , the analysis of the sn 2009bb nebular spectrum suggests that the failed grb detection is not imputable to a large angle between the line - of - sight and the grb beamed radiation .
therefore , if a grb was produced during the sn 2009bb explosion , it was below the threshold of the current generation of @xmath9-ray instruments . |
a remarkable and intimate connection between gravity and yang - mills amplitudes exist through the kawai - lewellen - tye ( klt ) relations @xcite . in the field theory limit ( @xmath0 ) these relations
take the form @xmath1 here @xmath2 , @xmath3 , @xmath4 are gravity and yang - mills amplitudes and @xmath5 is a specific function of kinematic invariants depending on the input yang - mills amplitudes .
the klt relations are a widely used tool for computing gravity amplitudes via a recycling of results for yang - mills amplitudes . for tree and loop amplitudes klt
can be combined with string theory inspired diagram rules as well as unitarity cut techniques @xcite .
recently , inspired by witten s work @xcite , we have seen remarkable progress in yang - mills amplitude computations ( for recent reviews see @xcite ) . via klt , this progress has fed into gravity yielding a much better understanding of amplitudes @xcite .
it is now clear that gravity tree amplitudes have additional simplicity in their expressions . in many cases
this simplicity appears very naturally from a string theory viewpoint @xcite .
gravity loop amplitudes also have simple forms and one - loop amplitudes in @xmath6 supergravity satisfy a no - triangle property @xcite .
this no - triangle property was first proven in @xcite .
no - triangle type of simplifications have been shown to hold even for pure gravity amplitudes @xcite and also at the level of multi - loop amplitudes .
specifically , @xmath6 supergravity amplitudes have been shown to satisfy very non - trivial cancellations at four - loop level @xcite ( see also @xcite ) .
interestingly , klt relations can be shown to hold regardless of the external matter content of the theory and in arbitrary dimension .
klt relations also hold for effective field theories of gravity @xcite .
we refer to ref .
@xcite for a recent review and for further references on klt and gravity amplitudes .
although the klt relations hold at the amplitude level in field theory , such relations have no natural framework at the lagrangian level .
it was however shown in a striking paper by bern and grant @xcite ( see also @xcite ) that one can construct graviton vertex interactions exclusively using the klt relations as well as qcd gluon amplitudes as input . in this paper
it is our aim to extend their analysis of graviton interactions to gravitinos and demonstrate how compact and factorized gravitino feynman rules can be fabricated . as in the paper by bern and grant
we will use the klt relations together with amplitude expressions from yang - mills theory to derive results for vertex rules .
as we are considering gravitino amplitudes we need to consider both gluon and fermion amplitudes .
all results will be demonstrated to be consistent with those derived via traditional methods .
tree amplitudes in gravity theories can be computed using the einstein - hilbert lagrangian which is given by @xmath7 we employ conventional metric @xmath8 and define @xmath9 . in the above equation @xmath10
denotes the determinant of the metric and @xmath11 is the ricci scalar .
the term @xmath12 contains possible matter interactions involving gravitons and other fields such as _ e.g. _ gravitinos . in order to derive vertex interactions one normally expands @xmath13
around flat space , ( _ i.e. _ @xmath14 ) , make a choice of gauge , and finally derive propagator and vertices from considering terms of order @xmath15 , @xmath16 , @xmath17 and various matter couplings respectively . in the text we will let @xmath18 , @xmath19 , @xmath10 and @xmath20 denote gravitons , gravitinos , gluons and gluinos .
we will use the spinor helicity formalism wherever convenient .
we define the antisymmetric tensor by @xmath21 and @xmath22 by @xmath23 .
all momenta in amplitudes are considered to be out - going .
we will define @xmath24 where @xmath25 and @xmath26 and @xmath27 is the charge conjugation matrix ( @xmath28 , @xmath29 , @xmath30 ) .
spinors are normalized so that @xmath31 at times we will use shorthand notation such as @xmath32=\langle p_a^+|p_b^-\rangle\,.\ ] ] all spinor products are anti - symmetric so that @xmath33 and they are all subject to the schouten identities @xmath34 we have @xmath35 and ( via a fierz rearrangement ) @xmath36\langle db\rangle \label{fierzing}\,.\ ] ] external gluinos can in the spinor helicity formalism be represented by either @xmath37 polarization vectors for gluons @xcite can be defined as @xmath38 here @xmath39 is a reference spinor ( we will at times suppress the reference spinors in equations to avoid unnecessary cluttering of expressions ) .
we will construct polarization tensors of the gravitino field via @xmath40 while the polarization tensors of the graviton field will be given by @xmath41 we will first consider more traditional ways of dealing with gravitino scattering ( see e.g. @xcite ) .
our starting point will be the following matter lagrangian @xmath42 @xmath43 this lagrangian governs the spin-@xmath44 gravitino field in curved space .
the covariant derivative is defined via @xmath45 $ ] and the spin connection @xmath46 is the usual one . to make a traditional computation of amplitudes with external gravitinos
, one can expand the kinetic part of the lagrangian ( here to lowest order in the graviton field ) .
discarding the antisymmetric vierbein field gives rise to the following lagrangian @xmath47 \psi_\sigma\partial_\beta h_{\rho\alpha}\label{lagrange}\\ & \hspace{3cm}+\frac{\kappa^2}{128 } \big(2\bar{\psi}_\mu\gamma_\nu\psi_\rho \bar{\psi}^\nu\gamma^\mu\psi^\rho+ \bar{\psi}_\mu\gamma_\nu\psi_\rho\bar{\psi}^\mu\gamma^\nu \psi^\rho-4\bar{\psi}_\mu\gamma^\mu\psi_\rho \bar{\psi}_\lambda\gamma^\lambda\psi^\rho\big)\ , .
\end{split}\ ] ] from this equation it is simple to find vertex factors @xcite .
_ e.g. _ if we have two gravitinos with momentum @xmath48 and @xmath49 ( lorentz indices @xmath50 and @xmath51 ) and a graviton leg with momentum @xmath39 ( lorentz indices @xmath52 ) we arrive at ( all factors of the coupling constant @xmath53 have been suppressed ) @xmath54(p+p')^\beta\label{nytrevertex } + [ \cancel{p}-\cancel{p}^\prime]\eta^{\mu\nu } \eta^{\alpha\beta}+\gamma^\alpha(p - p')^\beta \eta^{\mu\nu}\\&\hspace{7cm}-2\gamma^\alpha\eta^{\mu\beta } p^\nu+2\gamma^\alpha\eta^{\nu\beta}p^{\prime\mu}\big)\ , .
\end{split}\ ] ]
we will in this section use the klt relations to construct ( spin-@xmath44 ) gravitino feynman rules from yang - mills amplitudes involving gluons @xmath10 ( spin-1 ) and gluinos @xmath20 ( spin-@xmath55 ) .
klt relations valid up till five points are _
e.g. _ : @xmath56 @xmath57 @xmath58\ , .
\end{split}\ ] ] in these equations @xmath59 is a relative measure for the number of permutations of fermions on the right hand side of the equation as compared to the number of permutations on the left hand side , @xmath60 are generalized mandelstam variables , @xmath61 are gravity amplitudes ( with suppressed coupling constants ) and @xmath62 are color - ordered amplitudes defined in terms of the full amplitude as in @xcite ( see also @xcite ) .
the propagators and vertex factors for color - ordered gluon / gluino diagrams are shown in fig .
[ feynmanregler ] in feynman gauge .
( diagrams can have additional minus signs due to the fermionic nature of the gluinos .
for instance , in the amplitude of four gluinos an extra minus sign occurs when a fermionic line connects the first and the last external leg . )
we will use the following lagrangian for yang - mills theory , @xmath63 where @xmath64 $ ] is the covariant derivative , @xmath65 $ ] is the field tensor and @xmath66 is the coupling constant which will be suppressed throughout this paper .
@xmath67 denotes here the gluon field and @xmath68 the gluino field .
the generators of the group are normalized by @xmath69 and @xmath70=if^{abc}t^c$ ] . , title="fig:",width=642 ] inspired by the klt relations and the analysis by bern and grant @xcite , we will write a factorized three vertex for two gravitino one graviton scattering as follows @xmath71\times\bigg[\frac{i}{\sqrt{2}}\big((p - q)^\nu\eta^{\beta\mu}+ ( q - p')^\mu\eta^{\beta\nu}+(p'-p)^\beta\eta^{\mu\nu}\big)\bigg]\label{trevertex}\\ & \;=\ ; -i\,v^{\alpha}_{\widetilde{g}g\widetilde{g}}(p , q , p')\;v^{\mu,\beta,\nu}_{ggg}(p , q , p')\ , . \end{split}\ ] ] the vertex
is also shown diagrammatically below ( see fig . [ 3ver ] ) . ,
width=377 ] for the propagator of gravitons we can choose the same as in @xcite .
@xmath72 armed with a candidate for the gravitino three point vertex and a graviton propagator we can proceed to pursue what the four point contact interaction is . to find this
we start with the four point amplitude generated by the klt relation . from this amplitude
we subtract all contributions involving the three vertex .
what we are left with will form the four point contact vertex contribution .
the procedure is illustrated below ( see fig . [ figur4 ] ) .
, width=415 ] we begin by writing out the amplitude @xmath73 .
using klt we have @xmath74 in the above expression we can investigate the pole structure of the gluino part of the expression using the feynman rules .
we only have one diagram because the gluino - gluino - gluon vertex is helicity conserving .
thus we have ( we have for convenience suppressed the notion of the reference spinors in the polarization tensors and momenta in the vertex factors as well as in the propagators . ) @xmath75 one sees that the prefactor of @xmath76 from the klt relations cancels completely in this contribution .
now let us now focus our attention to the gluon part .
we will let the lorentz index of particle @xmath77 be @xmath78 .
there are three contributions corresponding to three distinct feynman graphs .
@xmath79 writing out @xmath73 using the above results we have @xmath80 the first term gives exactly a factorized gravitino vertex structure as suggested by eq .
( [ trevertex ] ) _ i.e. _ @xmath81 but the second term does not appear to do that . this can however be mended by applying the fierz rearrangement , @xmath82\langle db\rangle = -\langle p_a^{+}|\gamma_\mu|p_d^+ \rangle\langle p_c^{+}|\gamma^\mu|p_b^+\rangle\,.\ ] ] using this we can rearrange the gravitino amplitude so that it becomes : @xmath83 by subtracting from the amplitude the parts with the gravitino three point vertex and the propagator according to our substraction scheme ( see fig .
[ figur4 ] ) we have now recovered the structure of the gravitino contact term .
it should be kept in mind that there is no @xmath84 channel diagram because the gravitino - gravitino - graviton vertex is helicity conserving .
, width=548 ] the contact term is shown diagrammatically in fig .
[ kontakt ] . in the figure
the helicities are specified .
we can rewrite the contact term without reference to the helicities as follows ( using eq .
[ fierzing ] ) @xmath85 thus the contact term can be written ( again we have here suppressed factors of @xmath86 ) @xmath87 to leading order in @xmath88 , @xmath68 and @xmath18 one can thus instead of use the following interaction lagrangian : @xmath89 let us now turn to the amplitude of two gravitinos and two gravitons @xmath90 there are four diagrams to be considered .
they are shown below in fig .
[ firdia ] .
.[firdia],width=604 ] to calculate the diagrams ( a ) and ( b ) , we need in addition to the three vertex with one graviton leg ( see ) a propagator for the gravitino .
inspired by and the propagator used in @xcite we choose @xmath91 this propagator is shown diagrammatically in fig .
[ prophtilde ] . ,
width=302 ] in order to compute diagram ( c ) we need a graviton three vertex in addition to the already discussed feynman rules .
here we choose ( we have here suppressed factors of @xmath86 ) @xmath92 which is shown diagrammatically in fig .
[ 3verh ] : , width=415 ] it is convenient to split the diagrams @xmath93 , @xmath94 , @xmath95 and @xmath96 into a purely gluonic diagram and a mixed gluon / gluino diagram and let the poles be associated with the purely gluonic diagram ( we will define @xmath97 and @xmath98 and we have suppressed the notion of the momenta in the vertex factors as well as in the propagators for convenience ) : @xmath99 @xmath100 @xmath101 we will focus on the gluino parts of these expressions . we get ( see fig .
[ ggtildedia ] ) .
+ , width=642 ] + using this we can write @xmath93 , @xmath94 and @xmath95 as @xmath102 let us now sum the contributions from the three diagrams : @xmath103 by inspecting the terms in the parentheses , we notice that each of them consists of the pole diagrams from color - ordered gluon amplitudes .
hence if we add the gluon contact term @xmath104 and @xmath105 by hand we arrive at @xmath106 if we rewrite @xmath107 and @xmath108 in terms of the gluino / gluon diagrams we have @xmath109 we can now use that @xmath110 to rewrite @xmath111 thus we can read off the gravitino contact term as @xmath112 .
that correspond to the graviton / gravitino vertex rule ( we have again suppressed all factors of @xmath86 for convenience ) .
@xmath113\ , .
\end{split}\ ] ] at the level of the lagrangian this correspond to @xmath114 the vertex is shown diagrammatically below ( see fig . [ kontakthhtilde ] ) .
, width=604 ] finally , as a check of the feynman vertex rules of the previous section we turn to the amplitude of four gravitinos and one graviton .
there are five different feynman graph topologies as shown below ( see fig .
[ femdia ] ) .
, width=566 ] the diagrams of topology ( b ) , ( c ) , ( d ) and ( e ) can be created from the feynman rules found in the four point case .
contribution ( a ) requires the five point contact interaction . by a careful analysis of all contributions it turns out that no such term is necessary to get to the full amplitude .
thus the four gravitino one graviton contact vertex factor ( a ) is vanishing .
the vanishing of the five point interaction can also be understood from the perspective of the lagrangian given the manifest left / right mover separated structure of such a interaction term and the odd number of lorentz indices which subsequently have to be contracted with each other .
however , it is interesting that the klt inspired formalism yields this result with such ease .
similar arguments appear to hold for interaction terms in the lagrangian which have four gravitinos and any odd number of gravitons .
in this paper we have extended the analysis of bern and grant @xcite to gravitino scattering .
using yang - mills theory and the klt relations as the only input we have derived complete feynman rules for gravitino scattering verified until five - point scattering with one external graviton .
the feynman rules derived via klt have the useful property that they are simpler than results derived from a conventional analysis and they are manifestly factorized into a gluon and a gluino part .
it seems clear that the organizational principles induced by klt rearranges the lagrangian in useful and simpler ways than traditional gauge choices . using a klt inspired left - right separation of fields as a way to organize and symmetrize vertex interactions
is not limited to graviton and gravitino interactions . as a task for future research
it would be very interesting to further investigate how klt possibly could be used to simplify feynman rules and computations for many other types of matter .
we would like to thank p. h. damgaard and t. sndergaard for discussions and n. k. nielsen for clarifying some points concerning the article @xcite .
( nejbb ) is knud hjgaard assistant professor at the niels bohr international academy .
h. kawai , d. c. lewellen and s. h. h. tye , nucl .
b * 269 * , 1 ( 1986 ) .
h. kuijf , phys .
b * 211 * , 91 ( 1988 ) ; z. bern and d. c. dunbar , nucl .
b * 379 * , 562 ( 1992 ) ; z. bern , d. c. dunbar and t. shimada , phys .
b * 312 * , 277 ( 1993 ) [ hep - th/9307001 ] ; d. c. dunbar and p. s. norridge , nucl .
b * 433 * , 181 ( 1995 ) [ hep - th/9408014 ] ; d. c. dunbar and p. s. norridge , class .
* 14 * ( 1997 ) 351 [ hep - th/9512084 ] ; d. c. dunbar and n. w. p. turner , class .
* 20 * , 2293 ( 2003 ) [ hep - th/0212160 ] ; d. c. dunbar , b. julia , d. seminara and m. trigiante , jhep * 0001 * , 046 ( 2000 ) [ hep - th/9911158 ] .
z. bern , l. j. dixon , d. c. dunbar , m. perelstein and j. s. rozowsky , nucl .
b * 530 * , 401 ( 1998 ) [ hep - th/9802162 ] ; z. bern , l. j. dixon , m. perelstein and j. s. rozowsky , phys .
b * 444 * , 273 ( 1998 ) [ hep - th/9809160 ] ; z. bern , l. j. dixon , m. perelstein and j. s. rozowsky , nucl .
b * 546 * , 423 ( 1999 ) [ hep - th/9811140 ] ; z. bern , l. j. dixon , m. perelstein , d. c. dunbar and j. s. rozowsky , class .
* 17 * , 979 ( 2000 ) [ hep - th/9911194 ] .
z. bern and a. k. grant , phys .
b * 457 * , 23 ( 1999 ) [ hep - th/9904026 ] .
e. witten , commun .
phys . * 252 * , 189 ( 2004 ) [ hep - th/0312171 ] .
f. cachazo and p. svrcek , pos * rtn2005 * , 004 ( 2005 ) [ hep - th/0504194 ] .
z. bern , l. j. dixon and d. a. kosower , annals phys . * 322 * , 1587 ( 2007 ) [ 0704.2798 [ hep - ph ] ] . s. giombi , r. ricci , d. robles - llana and d. trancanelli , jhep * 0407 * , 059 ( 2004 ) [ hep - th/0405086 ] .
z. bern , n. e. j. bjerrum - bohr and d. c. dunbar , jhep * 0505 * , 056 ( 2005 ) [ hep - th/0501137 ] .
j. bedford , a. brandhuber , b. spence and g. travaglini , nucl .
b * 721 * , 98 ( 2005 ) [ hep - th/0502146 ] ; f. cachazo and p. svrek , hep - th/0502160 . n. e. j. bjerrum - bohr , d. c. dunbar and h. ita , phys .
b * 621 * , 183 ( 2005 ) [ hep - th/0503102 ] ; hep - th/0606268 ; hep - th/0608007 .
n. e. j. bjerrum - bohr , d. c. dunbar , h. ita , w. b. perkins and k. risager , jhep * 0601 * , 009 ( 2006 ) [ hep - th/0509016 ] .
n. e. j. bjerrum - bohr , d. c. dunbar , h. ita , w. b. perkins and k. risager , jhep * 0612 * , 072 ( 2006 ) [ hep - th/0610043 ] .
p. benincasa , c. boucher - veronneau and f. cachazo , jhep * 0711 * , 057 ( 2007 ) [ hep - th/0702032 ] .
z. bern , j. j. carrasco , d. forde , h. ita and h. johansson , phys .
d * 77 * , 025010 ( 2008 ) [ 0707.1035 [ hep - th ] ] .
s. ananth and s. theisen , phys .
b * 652 * , 128 ( 2007 ) [ 0706.1778 [ hep - th ] ] ; s. ananth , phys .
b * 664 * , 219 ( 2008 ) [ 0803.1494 [ hep - th ] ] ; s. ananth , fortsch .
phys . * 57 * , 857 ( 2009 ) [ 0902.3128 [ hep - th ] ] . h. elvang and d. z. freedman , jhep * 0805 * , 096 ( 2008 ) [ 0710.1270 [ hep - th ] ] ; m. bianchi , h. elvang and d. z. freedman , jhep * 0809 * , 063 ( 2008 ) [ 0805.0757 [ hep - th ] ] ; h. elvang , d. z. freedman and m. kiermaier , 0911.3169 [ hep - th ] .
a. hall , phys .
d * 77 * , 124004 ( 2008 ) [ 0803.0215 [ hep - th ] ]
. n. e. j. bjerrum - bohr and p. vanhove , jhep * 0804 * , 065 ( 2008 ) [ 0802.0868 [ hep - th ] ] ; fortsch .
* 56 * , 824 ( 2008 ) [ 0806.1726 [ hep - th ] ] .
n. arkani - hamed and j. kaplan , jhep * 0804 * , 076 ( 2008 ) [ 0801.2385 [ hep - th ] ] . n. e. j. bjerrum - bohr and p. vanhove , jhep * 0810 * , 006 ( 2008 ) [ 0805.3682 [ hep - th ] ]
. z. bern , j. j. m. carrasco and h. johansson , phys . rev .
d * 78 * , 085011 ( 2008 ) [ 0805.3993 [ hep - ph ] ] . n. arkani - hamed , f. cachazo and j. kaplan , 0808.1446 [ hep - th ] ; n. arkani - hamed , f. cachazo , c. cheung and j. kaplan , 0903.2110 [ hep - th ] .
s. badger , n. e. j. bjerrum - bohr and p. vanhove , jhep * 0902 * , 038 ( 2009 ) [ 0811.3405 [ hep - th ] ] .
p. katsaroumpas , b. spence and g. travaglini , jhep * 0908 * , 096 ( 2009 ) [ 0906.0521 [ hep - th ] ] .
z. bern , l. j. dixon and r. roiban , phys .
b * 644 * , 265 ( 2007 ) [ hep - th/0611086 ] ; h. johansson , d. a. kosower and r. roiban , phys .
lett . * 98 * , 161303 ( 2007 ) [ hep - th/0702112 ] ; z. bern , j. j. m. carrasco , l. j. dixon , h. johansson and r. roiban , phys .
d * 78 * , 105019 ( 2008 ) [ 0808.4112 [ hep - th ] ] ; z. bern , j. j. carrasco , l. j. dixon , h. johansson and r. roiban , phys .
lett . * 103 * , 081301 ( 2009 ) [ 0905.2326 [ hep - th ] ] . c. cheung and d. oconnell , jhep * 0907 * ,
075 ( 2009 ) [ 0902.0981 [ hep - th ] ] . n. e. j. bjerrum - bohr , p. h. damgaard and p. vanhove , phys .
lett . * 103 * , 161602 ( 2009 ) [ 0907.1425 [ hep - th ] ] . s. stieberger , 0907.2211 [ hep - th ] ; 0910.0180 [ hep - th ] .
r. boels , jhep * 1001 * , 010 ( 2010 ) [ 0908.0738 [ hep - th ] ] .
y. x. chen , y. j. du and q. ma , nucl .
b * 824 * , 314 ( 2010 ) [ 0901.1163 [ hep - th ] ] ; 1001.0060 [ hep - th ] .
m. b. green , j. g. russo and p. vanhove , jhep * 0702 * , 099 ( 2007 ) [ hep - th/0610299 ] ; m. b. green , j. g. russo and p. vanhove , phys .
lett . * 98 * , 131602 ( 2007 ) [ hep - th/0611273 ] ; n. berkovits , m. b. green , j. g. russo and p. vanhove , jhep * 0911 * , 063 ( 2009 ) [ 0908.1923 [ hep - th ] ] . s. weinberg , physicaa * 96 * , 327 ( 1979 ) ; j. f. donoghue , phys .
d * 50 * ( 1994 ) 3874 [ gr - qc/9405057 ] ; n. e. j. bjerrum - bohr , j. f. donoghue and b. r. holstein , phys .
d * 67 * , 084033 ( 2003 ) [ hep - th/0211072 ] ; phys .
d * 68 * , 084005 ( 2003 ) [ hep - th/0211071 ] .
z. bern , a. de freitas and h. l. wong , phys .
lett . * 84 * , 3531 ( 2000 ) [ hep - th/9912033 ] . n. e. j. bjerrum - bohr , phys
. lett .
b * 560 * , 98 ( 2003 ) [ hep - th/0302131 ] ; nucl . phys .
b * 673 * , 41 ( 2003 ) [ hep - th/0305062 ] ; n. e. j. bjerrum - bohr and k. risager , phys .
d * 70 * , 086011 ( 2004 ) [ hep - th/0407085 ] .
z. bern , living rev .
rel . * 5 * , 5 ( 2002 ) [ gr - qc/0206071 ] .
z. xu , d. h. zhang and l. chang , nucl .
b * 291 * , 392 ( 1987 ) .
w. rarita and j. schwinger , phys .
rev . * 60 * , 61 ( 1941 ) .
d. z. freedman , p. van nieuwenhuizen and s. ferrara , phys .
d * 13 * , 3214 ( 1976 ) .
s. ferrara , f. gliozzi , j. scherk and p. van nieuwenhuizen , nucl .
b * 117 * , 333 ( 1976 ) .
a. k. das and d. z. freedman , nucl .
b * 114 * , 271 ( 1976 ) .
m. t. grisaru , h. n. pendleton and p. van nieuwenhuizen , phys .
d * 15 * , 996 ( 1977 ) .
m. t. grisaru and h. n. pendleton , nucl .
b * 124 * , 81 ( 1977 ) .
n. k. nielsen , m. t. grisaru , h. romer and p. van nieuwenhuizen , nucl .
b * 140 * , 477 ( 1978 ) .
m. t. grisaru , d. z. freedman and p. van nieuwenhuizen , phys .
b * 71 * , 377 ( 1977 ) .
m. l. mangano and s. j. parke , phys .
* 200 * ( 1991 ) 301 [ hep - th/0509223 ] .
l. j. dixon , hep - ph/9601359 . | we fabricate gravitino vertex interactions using as only input on - shell yang - mills amplitudes and the kawai - lewellen - tye gauge theory / gravity relations .
a useful result of this analysis is simpler tree - level feynman rules for gravitino scattering than in traditional gauges .
all results are explicitly verified until five point scattering .
+ + |
cosmological observations imply the existence of dark matter ( dm ) to be the dominant component of cosmical matter@xcite .
its relic density has been determined precisely by the wmap experiment to be @xmath12@xcite and it is believed to be non - baryonic , cold , dissipationless and stable on time scales .
in addition to these , their physical properties , like making ups or deep natures , are still unknown . revealing the distribution and
the nature of dark matter is one of the most interesting challenges in the fields of both cosmology and particle physics .
many dark matter candidates have been proposed . weakly interacting massive particles ( wimps )
is the most compelling one among them .
many beyond the standard model ( bsm ) theories , such as supersymmetry@xcite , warped@xcite and universal@xcite extra dimensions or little higgs models@xcite , etc , predict good candidates for the wimps and for the cosmological requirements , i.e. , the wimp abundance is a natural consequence of the thermal history of the universe@xcite .
although well motivated , there is still no experimental evidence to support these theories
. it will be difficult to judge which theory is proper for dark matter .
even the observations of dark matter itself from future experiments may not provide enough information to distinguish underlying theories . in this case , a model independent studies of dark matter phenomenology using effective field theory ( eft ) can be particularly important .
various experiments are set up to hunt for the particle they make up and to reveal the nature of dark matter .
for instance , through direct detection ( dd ) experiments@xcite , we can search for scattering of dark matter particles from the galactic halo on detector nuclei . through indirect detection ( i d )
, we can detect dark matter particles by their annihilation into high energy standard model ( sm ) particles @xcite .
in addition , particle production through high energy accelerators will be another interesting way of dark matter hunting . in this case , the dark matter particles are expected to be detected as a missing component , or manifest as an excess of events showing an imbalance in momentum conservation .
there exist some experimental and theoretical studies include , i.e. , the visible radiation of a jet ( quark or gluon)@xcite , a photon@xcite , or a w / z boson decaying into leptons or hadronic jets@xcite plus the missing `` something '' .
the large hadron collider ( lhc ) at cern generates high energetic proton - proton ( @xmath13 ) collisions with a luminosity of @xmath14 and provides the opportunity to study very high energy physics . in such high energy , most attention
is usually paid to the central rapidity region where the most of the particles are produced and where the most of the high @xmath15 signal of new physics is expected .
indeed , the cdf collaboration has already observed such a kind of interesting phenomenon including the exclusive lepton pairs production @xcite , photon - photon production @xcite , dijet production @xcite and charmonium ( @xmath16 ) meson photoproduction @xcite , etc .
now , both the atlas and the cms collaborations have programs of forward physics , which are devoted to studies of high rapidity regions , with extra updated detectors located in a place nearly 100 - 400 m close to the interaction point @xcite .
technical details of the atlas forward physics ( afp ) projects can be found , for example , in refs.@xcite .
the physics program of this new instrumentation covers interesting topics like elastic scattering , diffraction , low - x qcd , central exclusive production ( cep ) , photon - photon ( @xmath17 ) and photo - proton ( @xmath0 ) interactions . dark matter searching will be an active topic and an important issue at the lhc@xcite . until now most works are concentrated on its searching through normal pp collision .
however , it will also be very interesting to see the status of dark matter searching in photon - photon ( @xmath17 ) and photo - proton ( @xmath0 ) interactions .
this is mainly due to the reason that photon interactions at the lhc are believed to be simple and clean from challenged backgrounds .
the study of photon interactions at the lhc might be a choice of extending the discovery bounds of dark matter as which that will be shown in our discussion . in this paper , we focus on the @xmath0 photoproduction of jet plus missing energy signal to set limits on the couplings of the fermionic dark matter to the quarks at the lhc via the main reaction @xmath1 .
we assume a typical lhc multipurpose forward detectors and work in a model independent eft framework .
paper is organized as follow : we build the calculation framework in section 2 including a brief introduction to @xmath0 collision and to the wimp production process we are interested in .
section 3 is arranged to present the numerical results and background analysis .
typically , the single diffractive ( sd ) production as background to @xmath0 productions is considered .
finally we summarize our conclusions in the last section .
a generic diagram for the photoproduction of @xmath18 at the cern lhc . ]
photoproduction is a class of processes in which one of the two interacting protons is not destroyed during the collision but survive into the final state with additional particle ( or particles ) state(s ) .
protons of this kind are named intact or forward protons . a generic diagram for the photoproduction denoted as @xmath18 is presented in fig.[rpexclusive ] .
the kinematics of a forward proton is often described by means of the reduced energy loss @xmath19 , which is also defined as the forward detector acceptance : @xmath20 where e is the initial energy of the beam and @xmath21 is the square of the centre of mass energy .
@xmath22 is the energy after the interaction and @xmath23 is the energy that the proton lost in the interaction . the possibility of adding forward proton detectors to both the atlas and the cms experiments has received quite some attention , since the forward proton tagging would provide a very clean environment for new physics searches .
different from usual @xmath13 deep inelastic scattering ( dis ) , incoming protons dissociate into partons , jets will be made from the proton remnants which create some ambiguities and make the new physics signal detection suffer from incredible backgrounds , @xmath17 and @xmath0 collisions can provide more clean environment . in this case , the quasi - real photons emit with very low virtuality from protons , leave the radiating proton intact , thus provide an extra experimental handle ( forward proton tagging ) to help reduce the backgrounds .
if both incoming and emitted protons remain intact , then it provides the @xmath17 collisions , which can be cleaner than the @xmath0 collisions .
however , @xmath0 collisions have higher energy and effective luminosity with respect to @xmath17 collisions
. deflected protons and their energy loss will be detected by the forward detectors with a very large pseudorapidity , while the remaining system will go to central detector .
photons emitted with small angles by the protons show a spectrum of virtuality @xmath24 and the energy @xmath25 .
this is described by the equivalent photon approximation ( epa)@xcite which differs from the point - like electron ( positron ) case by taking care of the electromagnetic form factors in the equivalent @xmath26 spectrum and effective @xmath26 luminosity : @xmath27\ ] ] with @xmath28 where @xmath29 is the fine - structure constant , e is the energy of the incoming proton beam , which is related to the quasi - real photon energy by @xmath30 .
@xmath31 is the mass of the proton and @xmath32 is the invariant mass of the final state .
@xmath33 = 7.78 is the magnetic moment of the proton .
@xmath34 and @xmath35 are functions of the electric and magnetic form factors given in the dipole approximation .
we denote the photoproduction processes in fig.[rpexclusive ] as @xmath36 with q = u , d , c , s , b and i , j , k , ... the final state particles .
the hadronic cross section at the lhc can be converted by integrating @xmath37 over the photon ( @xmath38 ) , gluon and quark ( @xmath39 ) spectra : @xmath40 where @xmath41 is the ratio between scattered quasi - real photons and incoming proton energy @xmath42 .
@xmath43 are its lower ( upper ) limits which means the forward detector acceptance satisfies @xmath44 .
@xmath45 is the momentum fraction of the proton momentum carried by the gluon ( quark ) .
the quantity @xmath46 is the effective center - of - mass system ( c.m.s . )
energy with @xmath47 .
@xmath21 mentioned above and @xmath32 is the total mass of the related final states .
@xmath48 is the jacobian determinant when transform the differentials from @xmath49 into @xmath50 .
@xmath51 represent the gluon ( quark ) parton density functions , @xmath52 is the factorization scale .
@xmath53 is the @xmath24 dependent relative luminosity spectrum present in eq.([epaformula ] ) .
@xmath54 is the maximum virtuality .
@xmath55 is the times of the spin - average factor , the color - average factor and the identical particle factor .
@xmath56 presents the squared n - particle matrix element and divided by the flux factor @xmath57 $ ] .
the n - body phase space differential @xmath58 and its integral @xmath59 depend only on @xmath60 and particle masses @xmath61 due to lorentz invariance : @xmath62 with i and j denoting the incident particles and k running over all outgoing particles .
a brief review of experimental prospects for studying photon induced interactions are summarized in ref.@xcite .
many other related phenomenological studies are summarized here : standard model productions@xcite , supersymmetry@xcite , extra dimensions@xcite , unparticle physics@xcite , top triangle moose model@xcite , gauge boson self - interactions @xcite , neutrino electromagnetic properties@xcite , the top quark physics@xcite and triplet higgs production@xcite , etc . in our study , we assume that the dark matter candidate is the only new particle which is singlet under the sm local symmetries , and all sm particles are singlets under the dark - sector symmetries . the interaction between dark matter and the sm particles are presumably effected by the exchange of some heavy mediators
whose nature we do not need to specify , but only assume that they are much heavier than the typical scales .
we specialize to the case of a spin-@xmath63 , dirac fermion and describe the interaction between dark matter and sm quarks accurately in terms of an effective field theory@xcite .
there consists of the sm lagrangian plus kinetic terms for the dark matter @xmath64 and a set of effective four fermion interactions between @xmath64 and the quarks q = u , d , s , c , b , t : @xmath65\end{aligned}\ ] ] where @xmath64 refers to the dark matter particle .
the characterizing parameters @xmath66 are the scales of the effective interactions where @xmath67 with m is the mass of the mediator , @xmath68 is the coupling to dark matter particle and @xmath69 is its coupling to sm quarks .
the first term in eq.([leftdm ] ) corresponds to the spin - independent scattering with vector coupling while the second term corresponds to the spin - dependent axial - vector coupling . here
we will consider interactions that yield a spin - independent scattering cross section and simplify our consideration to the vector coupling case .
similarly , our discussion can be easily extended to the other cases .
parton level feynman diagrams for @xmath1 include the s - channel contribution ( 1 ) and the t - channel contribution ( 2 ) .
here q = u , d , c , s , b , @xmath70 , @xmath71 , @xmath72 , @xmath73 and @xmath74 .
black blobs represent the @xmath75 couplings parameterized by eq.([leftdm ] ) . ] in the effective field theory , wimp - sm interactions lead to the direct production of wimps via the process @xmath76 through the normal pp collision mode at the colliders . in order to contain a visible trace
, one can study wimps produced together with hard jet(s ) , @xmath26(s ) or heavy gauge boson ( w , z ) and provide a trigger within hard scatterings .
different from normal pp collision , no direct production of @xmath76 appear at the @xmath0 collider .
the simplest production is @xmath77 . here
jet can not be @xmath26 or gluon but only light quarks .
the feynman diagrams can be found in fig.[dm ] .
we implement the lagrangian into feynarts and use the feynman diagram approach to provide the amplitudes . by inserting the amplitudes into eq.([totcrosssection ] ) we can obtain the total photon produced cross section for @xmath1 at the lhc .
we take the input parameters as @xmath78 , @xmath79 , @xmath80 , @xmath81 , @xmath82 , @xmath83@xcite and we have @xmath84 .
light fermion masses are neglected through the whole calculation .
the colliding energy in the proton - proton c.m.s .
system is assumed to be @xmath85 at the future lhc with luminosity taken to be a running parameter .
the factorization scale is chosen to be @xmath86 .
the strong coupling constant @xmath87 = 0.118 .
we use feynarts , formcalc and looptools ( ffl)@xcite packages to generate the amplitudes and perform the numerical calculations for both the signal and the background , with in - house modification that needed .
we adopt ct10@xcite pdf for the parton distributions during calculation and bases@xcite to do the phase space integration .
based on the forward proton detectors to be installed by the cms - totem and the atlas collaborations we choose the detected acceptances to be * cms - totem forward detectors with @xmath9 * cms - totem forward detectors with @xmath10 * afp - atlas forward detectors with @xmath11 which we simply refer to as @xmath6 , @xmath7 and @xmath8 , respectively .
such forward detected acceptances will force the proton intact together with the @xmath15 cuts of the related proton . in practically the detector
( i.e. , afp ) is independent of @xmath15 but much sensitive mainly to the @xmath19 value .
these acceptances with the following kinematical cuts will implicit the selected events to have an intact proton and to have been tagged by the forward detectors . during calculation
we use @xmath6 in default unless otherwise stated .
note that the gap survival probability ( @xmath88)@xcite in forward productions has some process dependence
. with early measurements of rapidity gap processes at the lhc , they will provide valuable information on @xmath88 .
we expect that @xmath89 . in our phenomenon
study the survival probability for @xmath0 collision production is chosen to be @xmath90 , while that for the single diffractive ( see following ) production is chosen to be @xmath91 . in all the results present in our work
, we impose a cut of pseudorapidity @xmath92 for the final state particles since central detectors of the atlas and cms have a pseudorapidity @xmath93 coverage 2.5 .
the general acceptance cuts for the events are : @xmath94 in order to reduce the backgrounds containing missing energy from processes where , for example , w bosons are produced , we remove such events for which a charged lepton has transverse momenta @xmath95 and is further separated from all jets by @xmath96@xcite , where @xmath97 is the separation in the rapidity - azimuth plane , @xmath98 are the transverse momentum of jets and leptons and @xmath99 the missing transverse momentum .
we stress here that cuts in eq.([cuts ] ) are the basic ones and might change later in our following discussion .
as can be seen , the studied topology of our signal therefore give rise to the @xmath100 signature characterized by one jet and a missing transverse momentum ( @xmath101 ) from the undetected dark matter pair production .
the dominant sm backgrounds consist of the following types : * @xmath0 collision of the sm production of z+jet where z decays into a pair of neutrinos : @xmath102 * @xmath0 collision of the sm w+jet associated production with the w decays into a neutrino and a charged lepton falls outside of the acceptance range of the detector : @xmath103 * single diffractive ( sd ) production of z+jet as main background to @xmath0 collision where z decays into a pair of neutrinos .
* single diffractive ( sd ) production of w+jet associated production with the w decays into a neutrino and the charged lepton falls outside of the acceptance range of the detector .
the first two in the list belong to the sm electroweak processes with some of their feynman diagrams presented in fig.[rqzw ] .
the cross section can be obtained by eq.([totcrosssection ] ) .
all the three generations of neutrino @xmath104 are taken into account .
the ckm matrix @xmath105 between different generation of quarks are omitted and assumed to be unit within the same generation .
feynman diagrams for @xmath0 collision of the sm production of z+jet where z decays into a pair of neutrinos and w+jet associated production with the w decays into a neutrino and a charged lepton falls outside the acceptance range of the detector .. ] now lets turn to the last two backgrounds listed ahead whose contributions come from the single diffractive productions .
sd production is indeed a kind of main background in photoproductions at the @xmath0 collider .
however , their contributions are usually omitted and not often well considered in some other phenomenological studies . in our paper , we take some efforts to include them and try to find out how important of their contributions will stand in background analysis at the @xmath0 collider . in order to provide a clear image to describe the sd production , in a logical clear way , we start from the description of double pomeron exchange ( dpe ) production as the main background of @xmath17 collision . in dpe ,
two colorless objects are emitted from both protons .
their partonic components are resolved and create a heavy mass object x ( @xmath106 ) in the central detector in the pomeron - pomeron interaction .
the event is characterized by two rapidity gaps between the central object and the protons . through the exchange of two pomerons , a dijet system , a diphoton system ,
a ww and zz pair , or a drell - yan pair can be created , for instance .
the dpe production cross section can be obtained within the factorized ingelman - schlein@xcite model . in the ingelman - schlein model
, one assumes that the pomeron has a well - defined partonic structure , and that the hard process takes place in the pomeron - proton / proton - pomeron ( single diffraction ) or pomeron - pomeron ( central diffraction ) processes .
these processes are described in terms of the proton `` diffractive '' parton distribution functions ( dpdfs ) .
similar to the parton distribution functions , the dpdfs are also obtained from deep inelastic scattering experiments@xcite .
the difference is that they contain a dependence on some additional variables which are used to describe the proton kinematics : the relative energy loss @xmath19 and four - momentum transfer t , which we introduce as @xmath107 where @xmath108 denotes either the quark or the gluon distributions with d refers to `` diffractive '' .
@xmath109 is the flux of pomerons and expressed as @xmath110 with @xmath111 and @xmath112 being the kinematic boundaries .
@xmath113 is the partonic structure of pomeron .
@xmath114 here is the fraction of the proton momentum carried by the pomeron corresponding to the relative energy loss @xmath19 .
and @xmath115 is the fraction of the pomeron momentum carried by the struck parton . both pomeron flux factors @xmath116 as well as quark
, gluon distributions in the pomeron @xmath117 were taken from the h1 collaboration analysis of diffractive structure function at hera@xcite .
therefore , the final forward detector acceptance @xmath19-dependent convolution integral for the dpe production is given by @xmath118 logically it will now easy to turn to sd production . a pomeron - proton or proton - pomeron interaction can be used to describe it and the total sd production cross section can be directly obtained by @xmath119 diffraction usually dominates for @xmath120 which means we can replace @xmath121 in eq.([eq.dpe ] ) or @xmath122 in eq.([eq.sd ] ) by 0.05 directly .
we check this and find the difference is quite small as expected . due to this reason , we can find that sd background are dominant for @xmath9 and @xmath11 while strongly suppressed for @xmath10 .
some feynman diagrams for single diffractive production of z+jet where z decays into a pair of neutrinos and w+jet associated production with the w decays into a neutrino and a charged lepton falls outside the acceptance range of the detector . ]
feynman diagrams for single diffractive production of z+jet with z decays into a pair of neutrinos and for that of w+jet associated production with the w decays into a neutrino and a charged lepton ( where the lepton falls outside the acceptance range of the detector ) , are presented in fig.[sdzw ] .
the double solid lines refer to the pomeron ( @xmath123 ) that come from the forward proton and going with a parton ( quarks or gluon ) emission from inside .
the other initial parton comes from the dissociate proton , described in terms of the normal proton pdf .
the other diagrams with the inverse of the pomeron - proton , which means replaced by proton - pomeron , are similar thus not shown in the figures .
final cross section of the sd productoin can than be obtain by eq.([eq.sd ] ) .
we add these productions into formcalc and perform our calculation also with the help of the ffl packages .
the signal and background differential cross sections of the transverse momentum of the jet ( @xmath124 ) , the invariant mass of the produced system ( @xmath32 ) and the @xmath125 distributions for @xmath1 . the basic kinematic cuts and survival probability factor are taken into account .
the forward detector acceptance is chosen to be @xmath9 . here
@xmath126 and @xmath127 and 200 gev.,title="fig : " ] the signal and background differential cross sections of the transverse momentum of the jet ( @xmath124 ) , the invariant mass of the produced system ( @xmath32 ) and the @xmath125 distributions for @xmath1 .
the basic kinematic cuts and survival probability factor are taken into account .
the forward detector acceptance is chosen to be @xmath9 . here
@xmath126 and @xmath127 and 200 gev.,title="fig : " ] the signal and background differential cross sections of the transverse momentum of the jet ( @xmath124 ) , the invariant mass of the produced system ( @xmath32 ) and the @xmath125 distributions for @xmath1 .
the basic kinematic cuts and survival probability factor are taken into account .
the forward detector acceptance is chosen to be @xmath9 . here
@xmath126 and @xmath127 and 200 gev.,title="fig : " ] in fig.[distri ] we present the differential cross sections of the transverse momentum of the jet ( @xmath124 ) , the invariant mass of the produced system ( @xmath32 ) and the @xmath125 distributions where @xmath128 . for the signal ,
we have chosen a wimp mass of 10 gev and the effective scale @xmath127 and 200 gev .
the basic kinematic cuts in eq.([cuts ] ) and the survival probability factor @xmath129 and @xmath130 are taken into account .
the forward detector acceptance is chosen to be @xmath9 .
different sub - contributions of the backgrounds including @xmath0 electroweak productions and sd productions are presented .
their total contributions are depicted by the solid lines .
compare these background contributions we can see for @xmath9 , the dominant background indeed come from sd z+jet production .
the smallest one is the photon produced w+jet production .
the @xmath0 z+jet production and sd w+jet production are comparable and not deviate far from each other .
no matter which one , all the backgrounds are dominant over the low @xmath124 region and reduced rapidly through the high @xmath124 region .
the same behavior can be found for distributions of @xmath32 and @xmath125 .
the dashed and dotted lines present the signal with @xmath131 and @xmath132 , respectively .
all the @xmath124 , @xmath32 and @xmath125 distributions are enhanced by the dark matter effects at larger or higher regions .
this potential of distributions clearly provide very simple and effective way to separate the signal from the backgrounds .
this is easily and primarily realized by adopting large @xmath124 cuts .
just to remind here that in fig.[distri ] we choose the basic kinematic cuts where @xmath133 . in order to see how the cross sections for signal and background depend on the final transverse momentum jet ( @xmath134 ) cuts
, we present this dependence in tab.[ssx1 ] with typical cuts of @xmath133 , @xmath135 , @xmath136 and @xmath137 , respectively .
we can see that larger @xmath134 cut can reduce the background cross section essentially while make the signal reduce slightly . as can be found that the larger @xmath134 cut can improve the quantity of the signal over background ratio ( s / b ) , which implies a result that is more robust against systematic uncertainties .
we also calculate the statistical significance ( ss ) for the signal and background on different values of @xmath134 cuts in the same table for @xmath138 with the following formula@xcite : @xmath139 } \end{aligned}\ ] ] where s and b are the numbers of signal and background events , respectively .
@xmath140 presents the luminosity of future 14 tev lhc where we take the value of 5 and 20 @xmath141 .
we can see the statistical significance of the dark matter effect is large and might be detected . in the following calculation , we apply @xmath142 for @xmath9 . the same table but different values is presented for @xmath143 in tab.[ssx2 ] .
in this case , we apply @xmath144 . from our former studies@xcite
we know that the production behavior for @xmath145 is very close to that of @xmath9 .
this has also been calculated and confirmed .
results are not shown here to avoid duplicate .
similarly here we adopt @xmath142 .
c c c c c + @xmath146[pb ] & @xmath147 & @xmath148 & @xmath149 & @xmath150 + @xmath151 & 3.2725 & 4.0933 @xmath152 & 5.1713 @xmath153 & 1.1613@xmath153 + @xmath154 & 9.1687 & 2.9393 & 1.4921 & 8.3580 @xmath155 + @xmath156 & 1.1323 @xmath155 & 3.6288 @xmath152 & 1.8421 @xmath152 & 1.0318 @xmath152 + @xmath157 & 4.40079 & 11.2813 & 13.1863 & 12.6422 + @xmath158 & 8.80158 & 22.5625 & 26.3726 &
25.2844 + c c c c c + @xmath146[pb ] & @xmath147 & @xmath148 & @xmath149 & @xmath150 + @xmath151 & 2.8022@xmath152 & 1.6099@xmath153 & 2.4527@xmath159 & 7.2497@xmath160 + @xmath154 & 4.3063 & 1.9020 & 1.1068 & 6.8305 @xmath155 + @xmath156 & 5.3164 @xmath152 & 2.3481 @xmath152&1.3664@xmath152 & 8.4327@xmath153 + @xmath157 & 18.2204 & 21.3139 & 20.6161 & 17.9147 + @xmath158 & 36.4408 & 42.6277 & 41.2322 & 35.8294 + the total cross sections of signal process @xmath1 as functions of the dark matter effective scale @xmath161 and three forward detector acceptance regions : @xmath9 , @xmath10 and @xmath11 .
notice here we choose @xmath162 for @xmath6 , @xmath7 and @xmath8 , respectively .
the dark matter particle mass is @xmath163 . ] in fig.[mxcrosssection ] , we present the signal cross sections of @xmath1 as functions of the dark matter scale @xmath161 and three forward detector acceptance : @xmath9 , @xmath10 and @xmath11 .
solid , dashed and dotted lines correspond to @xmath6 , @xmath7 and @xmath8 , respectively .
their cross sections dropped rapidly as @xmath161 become larger . for results of @xmath161 and @xmath164 , their contributions are indeed very close to each other .
that s why we only focus our study on @xmath165 .
compare different acceptance regions , though they have almost the same features , @xmath6 and @xmath8 do not differ much from each other while both of them are much larger than cross section of @xmath7 . just to remind that here we choose @xmath166 for @xmath6 , @xmath7 and @xmath8 , respectively .
that s why cross section of @xmath6 is smaller than that of @xmath7 , otherwise if with the same value of @xmath134 it will inverse .
the sm backgrounds for the main reactions are @xmath167 for @xmath9 , @xmath168 for @xmath10 and @xmath169 for @xmath11 .
these numbers will be used in our following data analysis . from this point , we see deviation of the dark matter production from the backgrounds are obvious which might detectable from future experiments .
@xmath170 and @xmath2 bounds for the dark matter effective scale in the @xmath171 parameter space , and @xmath170 and @xmath2 limit of exclusive boundary where effective field theory may not valid .
the dark matter particle mass is @xmath163.,title="fig : " ] @xmath170 and @xmath2 bounds for the dark matter effective scale in the @xmath171 parameter space , and @xmath170 and @xmath2 limit of exclusive boundary where effective field theory may not valid .
the dark matter particle mass is @xmath163.,title="fig : " ] @xmath170 and @xmath2 bounds for the dark matter effective scale in the @xmath171 parameter space , and @xmath170 and @xmath2 limit of exclusive boundary where effective field theory may not valid .
the dark matter particle mass is @xmath163.,title="fig : " ] now we turn to analyze the potential of the @xmath0 collision at the high energy lhc to probe the dark matter signal through photoproduction of @xmath172 .
we define the @xmath170 ( @xmath2 ) discovery significance as @xmath173 where @xmath174 is the integrated luminosity of the @xmath0 option of the lhc . in the @xmath171 parameter space
, we present the plots for the @xmath170 and @xmath2 deviations of the signal from the total backgrounds .
we display our results for the forward detector acceptance chosen to be @xmath9 , @xmath10 and @xmath11 in fig.[mxbound ] , the first , the second and the third panel , respectively .
the center of mass energies of the lhc collider is @xmath175 = 14 tev .
concerning the criteria 5@xmath146 deviation of signal from background , with an integrated luminosity @xmath176 , @xmath1 production through jet photoproduction can probe the dark matter effects up to @xmath161 equal 602.7 , 748.1 and 510.2 gev for @xmath6 , @xmath7 and @xmath8 . with an integrated luminosity @xmath4
one can probe the dark matter effects up to the mass scale @xmath161 equal 665.5 , 808.9 , 564.0 gev for @xmath6 , @xmath7 and @xmath8 , respectively . here
we choose @xmath163 , much less than the scale @xmath177 .
we use solid and dashed lines to present the @xmath170 and @xmath2 boundary , respectively .
compare these three forward detector acceptances @xmath7 is the most efficient one to detect the dark matter effects .
this is mainly because @xmath178 belong to the region where single diffractive productions are strongly suppressed , leaving electroweak photoproductions the main background which are indeed quite small with our kinematic cuts , thus lead to a better signal over background ratio . assuming a 3@xmath146 deviation of signal from background , for @xmath7 , @xmath0 collision of @xmath172 can probe @xmath177 up to @xmath179 , a factor of 2.5 above the best current bounds from lep-2@xcite .
our bounds are comparable with the phenomenological study at a 250 gev ilc , however , worse than the polarized beams case@xcite .
similar studies at the normal pp collider can be found , i.e. , in ref.@xcite where the boundary of dark matter mass scale can be probe up to 450 gev at 14 tev lhc with the a luminosity of 200 @xmath141 .
further more , cms and atlas have both carried out the cut - and - count based monojet@xcite and monophoton@xcite analyses ( using @xmath180 of data ) .
typically , for the searches through monophoton channel , in the case of d5 spin - independent operator , values of scale below 585 gev and 156 gev are excluded at @xmath181 c.l . for @xmath182
equal to 1 gev and 1.3 tev , respectively .
our results show that with the same value of luminosity , the bound obtained is comparable with that obtained by the forward detector acceptance @xmath183 , while smaller than that obtained by @xmath7 .
it should be noted that the contact operator approximation may not be valid if the energy scale of the operator is smaller than the energy scale of the collider process .
the bounds we have already obtained may only be somewhat heuristic@xcite .
this means that an experiment can exclude such a range of @xmath184 . in our case
the limit is relevant for collider dark matter searches , thus @xmath185 here is the invariant mass of the invisibles . in fig.[mxbound ] we also present the @xmath170 and @xmath2 limit boundary ( with dotted and short dashed curves , respectively ) of the range that should be excluded where the effective field theory may not valid . in fig.[mxbound ]
, we also present the bounds obtained when the luminosity becomes larger than @xmath186 , see , up to @xmath187 .
however , we should mention here that as the luminosity become larger , identify the signal under the high pileup running conditions will be challenge : the hadronic background of multiple p - p interactions will be so large that any @xmath0 process will be completely swamped .
this can be a drawback of @xmath0 productions .
even though , purpose of our study does not mean to argue that @xmath0 production are better or not to probe dark matter than the other colliders .
instead , the main purpose is to show the potential of @xmath0 productions on the dark matter searching . by applying the dark matter searching on the @xmath0 production with forward proton tagging
we can have at least a general image of the dark matter searching through @xmath0 collision after the running of lhc but before the building of future @xmath188 colliders .
moreover , the production channel we consider here ( @xmath189 ) is related by crossing symmetry to the hard process considered in monophoton dark matter searches ( @xmath190 ) .
the advantage of @xmath189 production is the reduction in background associated with tagging the intact proton . while the advantage of the monophoton production would be that
: a photon in the final state may be cleaner than a jet , and one is more likely to get a @xmath191 from the initial proton than a photon .
it will also be interesting to study a monophoton search in which the initial q or @xmath192 arises from pomerons which would have a potentially larger production cross - section , and a cleaner final state .
a similar study has been presented in ref.@xcite where they concentrate on the standard model z boson production .
it can also be extended to the study of new physics , i.e. , dark matter searches .
a detailed discussion on kinds of these productions might be beyond the scope of this work , we address here the possibility and interesting to study them , and will work on this in a parallel paper .
dark matter searches at the high energy colliders have received a lot of attention .
the study of wimps in a model independent effective field theory ( eft ) approach is of particularly interesting . even if a process can be traced back to a definite set of operators , it is rarely the case that a particular collider signature can be traced back to a unique process .
for this reason many different , complementary measurements on different collider modes and production channels are usually required to uncover the underlying new physics .
lhc provides opportunities to open new fields of the study on the @xmath17 and @xmath0 collisions with very high energy but very low backgrounds thanks to the forward proton projects .
attention can be paid now to physics in the central rapidity region where the most of the high @xmath15 signal of new physics is expected . in this paper , we investigate the @xmath0 photoproduction of jet plus missing energy signal to set limits on the couplings of the fermionic dark matter to the quarks at the lhc via the main reaction @xmath1 .
we assume a typical lhc multipurpose forward detectors and work in a model independent eft framework .
typically , when we do the background analysis , we also include their corresponding single diffractive ( sd ) productions .
our result shows that by requiring a @xmath2 ( @xmath3 ) signal deviation , with an integrated luminosity of @xmath4 , the lower bounds of the dark matter mass scale can be probed up to @xmath161 equal 665.5 , 808.9 , 564.0 gev for the forward detector acceptances @xmath193 , respectively , where @xmath9 , @xmath10 and @xmath11 . assuming a 3@xmath146 deviation of signal from background , for @xmath7
, @xmath0 collision of @xmath172 can probe @xmath177 up to @xmath179 , a factor of 2.5 above the best current bounds from lep-2 can comparable with similar studies at the normal pp colliders . by applying the dark matter searching on the @xmath0 production with forward proton tagging we present a general image of the dark matter searching at the @xmath0 collision after the running of lhc but before the building of future @xmath188 colliders .
the author thanks fawzi boudjema for his kindness to provide invaluable comments , thanks mao song for useful discussions . project supported by the national natural science foundation of china ( no . 11205070 ) , shandong province natural science foundation ( no .
zr2012aq017 ) and by the fundamental research funds for the central universities ( no . dut13rc(3)30 ) .
h. goldberg , phys .
50 ( 1983 ) 1419 ; j. r. ellis , j. s. hagelin , d. v. nanopoulos , k. a. olive and m. srednicki , nucl .
b 238 ( 1984 ) 453 ; g. l. kane , c. f. kolda , l. roszkowski and j. d. wells , _ study of constrained minimal supersymmetry _ , phys . rev .
d 49 ( 1994 ) 6173 - 6210 [ arxiv : hep - ph/9312272 ] .
k. agashe and g. servant , _ warped unification , proton stability and dark matter _ , phys .
93 ( 2004 ) 231805 , [ arxiv : hep - ph/0403143 ] ; _ baryon number in warped guts : model building and ( dark matter related ) phenomenology _ , jcap 0502 ( 2005 ) 002 , [ arxiv : hep - ph/0411254 ] ; k. agashe , a. falkowski , i. low and g. servant , _ kk parity in warped extra dimension _ , jhep 0804 ( 2008 ) 027 , [ arxiv:0712.2455 ] ; g. panico , e. ponton , j. santiago and m. serone , _ dark matter and electroweak symmetry breaking in models with warped extra dimensions _ ,
phys . rev .
d 77 ( 2008 ) 115012 , [ arxiv:0801.1645 ] .
t. appelquist , h. c. cheng and b. a. dobrescu , _
bounds on universal extra dimensions _ , phys .
d 64 ( 2001 ) 035002 , [ arxiv : hep - ph/0012100 ] ; g. servant and t. m. p. tait , _ is the lightest kaluza - klein particle a viable dark matter candidate ? _ , nucl . phys .
b 650 ( 2003 ) 391 , [ arxiv : hep - ph/0206071 ] ; h. c. cheng , j. l. feng and k. t. matchev , _ kaluza - klein dark matter _ , phys .
( 2002 ) 211301 , [ arxiv : hep - ph/0207125 ] ; d. hooper and s. profumo , _ dark matter and collider phenomenology of universal extra dimensions _
( 2007 ) 29 - 115 , [ arxiv : hep - ph/0701197 ] ; b. a. dobrescu , d. hooper , k. kong and r. mahbubani , _ spinless photon dark matter from two universal extra dimensions _
, jcap 0710 ( 2007 ) 012 , [ arxiv:0706.3409 ] . n. arkani - hamed , a. g. cohen and h. georgi , _ electroweak symmetry breaking from dimensional deconstruction _ ,
b 513 ( 2001 ) 232 - 240 , [ arxiv : hep - ph/0105239 ] ; h. c. cheng and i. low , _ tev symmetry and the little hierarchy problem _ , jhep 0309 ( 2003 ) 051 , [ arxiv : hep - ph/0308199 ] ; a. birkedal , a. noble , m. perelstein and a. spray , _ little higgs dark matter _
, phys . rev .
d 74 ( 2006 ) 035002 , [ arxiv : hep - ph/0603077 ] .
feng and j. kumar , _ the wimpless miracle : dark - matter particles without weak - scale masses or weak interactions _ , phys .
( 2008 ) 231301 , [ arxiv:0803.4196 ] ; j .-
feng , h. tu and h .- b .
yu , _ thermal relics in hidden sectors _ , jcap 0810 ( 2008 ) 043 , [ arxiv:0808.2318 ] .
d. s. akerib , et al . , ( lux collaboration ) , _ first results from the lux dark matter experiment at the sanford underground research facility _ , phys
( 2014 ) 091303 , [ arxiv:1310.8214 ] ; e. aprile , et al . , ( xenon100 collaboration ) , _ dark matter results from 225 live days of xenon100 data _ , phys .
rev . lett . 109
( 2012 ) 181301 , [ arxiv:1207.5988 ] .
m. felizardo , t. a. girard , t. morlat , a. c. fernandes , a. r. ramos , j. g. marques , a. kling and j. puibasset , et al .
, _ final analysis and results of the phase ii simple dark matter search _
( 2012 ) 201302 , [ arxiv:1106.3014 ] ; e. vazquez - jauregui , ( coupp collaboration ) , _ coupp : bubble chambers for dark matter detection _ , in proceedings of the 48th rencontres de moriond on very high energy phenomena in the universe .
cdf collaboration , _ a search for dark matter in events with one jet and missing transverse energy in p@xmath194 collisions at @xmath175 = 1.96 tev _
lett 108 ( 2012 ) 211804 , [ arxiv:1203.0742 ] .
g. aad , et al . , ( atlas collaboration ) , _ search for dark matter candidates and large extra dimensions in events with a jet and missing transverse momentum with the atlas detector _ , jhep 1304 ( 2013 ) 075 , [ arxiv:1210.4491 ] .
g. aad , et al . , ( atlas collaboration ) , _ search for dark matter candidates and large extra dimensions in events with a photon and missing transverse momentum in pp collision data at @xmath175 = 7 tev with the atlas detector _
lett 110 ( 2013 ) 011802 , [ arxiv:1209.4625 ] .
s. chatrchyan , et al . , ( cms collaboration ) , _
search for dark matter and large extra dimensions in pp collisions yielding a photon and missing transverse energy _ ,
( 2012 ) 261803 , [ arxiv:1204.0821 ] . the atlas collaboration , _ search for dark matter pair production in events with a hadronically decaying w or z boson and missing transverse momentum in pp collision data at @xmath175 = 8 tev with the atlas detector _ , atlas - conf-2013 - 073
g. aad , et al . , ( atlas collaboration ) , _
search for dark matter in events with a hadronically decaying w or z boson and missing transverse momentum in pp collisions at @xmath175=8 tev with the atlas detector _ ,
( 2014 ) 041802 , [ arxiv:1309.4017 ] .
t. aaltonen , et al .
, ( cdf collaboration ) , _
search for exclusive z - boson production and observation of high - mass @xmath195 events in @xmath196 collisions at @xmath197 tev _ , phys .
102 ( 2009 ) 222002 , [ arxiv:0902.2816 ] .
t. aaltonen , et al . , ( cdf collaboration ) , _ search for exclusive @xmath17 production in hadron - hadron collisions _ , phys
99 ( 2007 ) 242002 , [ arxiv:0707.2374 ] ; _ observation of exclusive @xmath17 production in @xmath196 collisions at @xmath197 tev _ ,
phys.rev.lett . 108
( 2012 ) 081801 , [ arxiv:1112.0858 ] .
t. aaltonen , et al . , ( cdf collaboration ) , _ observation of exclusive charmonium production and @xmath199 in @xmath196 collisions at @xmath197 tev _ , phys . rev
102 ( 2009 ) 242001 , [ arxiv:0902.1271 ] . c. royon , ( rp220 collaboration ) , _ project to install roman pot detectors at 220 m in atlas _ , [ arxiv:0706.1796 ] ; m.g .
albrow , et al . , _
fp420 : an r@xmath200d proposal to investigate the feasibility of installing proton tagging detectors in the 420-m region at lhc _ , cern - lhcc-2005 - 025 ( 2005 ) ; b.e .
cox , ( fp420 r and d collaboration ) , _ the fp420 r@xmath200d project at the lhc _ , [ arxiv : hep - ph/0609209 ] .
atlas collaboration , _
search for new phenomena in monojet plus missing transverse momentum final states using 10@xmath201 of pp collisions at @xmath175 = 8 tev with the atlas detector at the lhc _ , atlas - conf-2012 - 147 .
v. m. budnev , i. f. ginzburg , g. v. meledin , v. g. serbo , _ the two - photon particle production mechanism .
physical problems . applications .
equivalent photon approximation _
, phys . rep . 15
( 1975 ) 181 ; g. baur , k. hencken , d. trautmann , s. sadowsky and y. kharlov , _
coherent @xmath17 and @xmath202 interactions in very peripheral collisions at relativistic ion colliders _ , phys . rep .
364 ( 2002 ) 359 , [ arxiv : hep - ph/0112211 ] ; k. piotrzkowski , _ tagging two - photon production at the cern large hadron collider _ ,
d 63 ( 2001 ) 071502 , [ hep - ex/0009065 ] .
n. schul , k. piotrzkowski , _ detection of two - photon exclusive production of supersymmetric pairs at the lhc _ ,
179 - 180 ( 2008 ) 289 - 297 , [ arxiv:0806.1097 ] ; k. piotrzkowski , n. schul , _ two - photon exclusive production of supersymmetric pairs at the lhc _ , aip conf.proc . 1200
( 2010 ) 434 - 437 , [ arxiv:0910.0202 ] .
s. heinemeyer , v. a. khoze , m. g. ryskin , w. j. stirling , m. tasevsky , g. weiglein , _ studying the mssm higgs sector by forward proton tagging at the lhc _ , eur .
j. c 53 ( 2008 ) 231 - 256 , [ arxiv:0708.3052 ] ; _ central exclusive diffractive mssm higgs - boson production at the lhc _ , j. phys .
110 ( 2008 ) 072016 , [ arxiv:0801.1974 ] ; s. heinemeyer , v.a .
khoze , m.g .
ryskin , m. tasevsky , g. weiglein , _ bsm higgs physics in the exclusive forward proton mode at the lhc _
j. c 71 ( 2011 ) 1649 , [ arxiv:1012.5007 ] ; marek tasevsky , _ exclusive mssm higgs production at the lhc after run i _ , eur .
j. c 73 ( 2013 ) 2672 , [ arxiv:1309.7772 ] . c. royon , e. chapon , o. kepka , _ anomalous trilinear and quartic ww@xmath26 , ww@xmath17 , zz@xmath26 and zz@xmath17 couplings in photon induced processes at the lhc _
, pos eps - hep2009 ( 2009 ) 380 , [ arxiv:0909.5237 ] .
e. chapon , c. royon , o. kepka , _ anomalous quartic @xmath206 , @xmath207 , and trilinear @xmath205 couplings in two - photon processes at high luminosity at the lhc _ , phys .
d 81 ( 2010 ) 074003 , [ arxiv:0912.5161 ] .
m. beltran , d. hooper , e.w . kolb and z.c .
krusberg , _ deducing the nature of dark matter from direct and indirect detection experiments in the absence of collider signatures of new physics _ , phys .
d 80 ( 2009 ) 043509 , [ arxiv:0808.3384 ] ; hao zhang , qing - hong cao , chuan - ren chen , chong sheng li , _ effective dark matter model : relic density , cdms ii , fermi lat and lhc _ , jhep 1108 ( 2011 ) 018 , [ arxiv:0912.4511 ] ; j. goodman , m. ibe , a. rajaraman , w. shepherd , t.m.p . tait and h .- b .
yu , _ constraints on light majorana dark matter from colliders _ , phys . lett .
b 695 ( 2011 ) 185 , [ arxiv:1005.1286 ] . t. hahn , _ automatic loop calculations with feynarts , formcalc , and looptools _ , nucl . phys .
89 ( 2000 ) 231 - 236 , [ arxiv : hep - ph/0005029 ] .
s. agrawal , t. hahn , e. mirabella , _ formcalc 7 _ , j. phys .
ser . 368 ( 2012 ) 012054 , [ arxiv:1112.0124 ] .
marco guzzi , pavel nadolsky , edmond berger , hung - liang lai , fredrick olness , c .- p .
ct10 parton distributions and other developments in the global qcd analysis _ , smu - hep-10 - 11 , [ arxiv:1101.0561 ] .
s. kawabata , _ a new version of the multi - dimensional integration and event generation package bases / spring _ , comp .
commun 88 ( 1995 ) 309 - 326 ; f. yuasa , d. perret - gallix , s. kawabata , t. ishikawa , _ pvm - grace _ , nucl .
instrum . meth .
a389 ( 1997 ) 77 - 80 .
a. cisek , w. schafer , a. szczurek , _ production of @xmath215 bosons with rapidity gaps : exclusive photoproduction in @xmath26p and pp collisions and inclusive double diffractive @xmath216 _ , phys .
d 80 ( 2009 ) 074013 , [ arxiv:0906.1739 ] . | in this paper , we investigate the @xmath0 photoproduction of jet plus missing energy signal to set limits on the couplings of the fermionic dark matter to the quarks at the lhc via the main reaction @xmath1 .
we assume a typical lhc multipurpose forward detectors and work in a model independent effective field theory framework .
typically , when we do the background analysis , we also include their corresponding single diffractive ( sd ) productions .
our result shows that by requiring a @xmath2 ( @xmath3 ) signal deviation , with an integrated luminosity of @xmath4 , the lower bounds of wimp masses scale can be detected up to @xmath5 equal 665.5 , 808.9 and 564.0 gev for the forward detector acceptances @xmath6 , @xmath7 , and @xmath8 , respectively , where @xmath9 , @xmath10 and @xmath11 .
* keywords * : dark matter , forward detector , large hadron collider + * pacs numbers * : 12.60.-i , 14.70.bh , 95.35.+d |
redshift surveys have shown that the clustering properties of galaxies strongly depend on their luminosity , color and morphological ( or spectral ) type ( e.g. @xcite ; @xcite ) .
this indicates that galaxies do not perfectly trace the distribution of the underlying dark matter , a phenomenon commonly referred to as ` galaxy biasing ' .
its origin lies in the details of the galaxy formation process which is shaped by the interplay between complex hydrodynamical and radiative processes together with the dark - matter driven formation of the large - scale structure .
attempts to infer cosmological parameters from galaxy clustering studies are severely hampered by galaxy biasing .
a number of theoretical arguments and the outcome of numerical simulations both suggest that , on sufficiently large scales , the power spectra of galaxies and matter should be proportional to each other : @xmath5 where the linear bias factor @xmath6 depends on galaxy type but is generally scale independent ( e.g. @xcite ; @xcite ) .
similarly , to model higher - order statistics , such as the galaxy bispectrum , it is generally assumed that galaxy biasing is a local process such that @xmath7 where @xmath8 and @xmath9 are the ( smoothed ) galaxy and dark - matter density contrast , respectively @xcite .
however , the reliance of these phenomenological approximations limits cosmological studies to very large scales whereas data with better signal - to - noise ratio are already available on much smaller scales .
moreover , future studies of baryonic acoustic oscillations ( e.g. @xcite ; @xcite ) will require measurements of the matter power - spectrum with percent or even sub - percent accuracy in order to shed new light on the source of cosmic acceleration .
understanding and controlling the effects of galaxy biasing with this precision will be challenging .
all this provides a very strong motivation for developing more accurate ( and physically driven ) models of galaxy biasing .
a number of authors have used the power spectrum statistics to explore the scale dependence of galaxy biasing based on numerical simulations ( @xcite ; @xcite ; @xcite ; @xcite ; @xcite ; @xcite ) or analytical calculations ( @xcite ; @xcite ; @xcite ; @xcite ) stemming from either perturbation theory or the halo model for the large - scale structure ( see @xcite for a review ) .
the general picture is that galaxy biasing is expected to be scale dependent ( i.e. @xmath10 ) and the functional form of @xmath11 can sensibly depend on the selected tracer of the large - scale structure . since galaxies are expected to form within dark - matter haloes , understanding the clustering properties of the haloes is a key step to accurately model galaxy biasing .
this is a much simpler problem , considering that dark - matter haloes form under the sole action of gravity .
it is in fact expected that long - wavelength density fluctuations modulate halo formation by modifying the collapse time of localized short - wavelength density peaks ( @xcite ; @xcite ) .
this argument ( known as the peak - background split ) predicts that , on large scales , the halo overdensity @xmath12 where the bias coefficient @xmath13 varies with the halo mass @xcite . the numerical value of the bias coefficient is determined by two different occurrences : first , haloes form out of highly biased regions in the linear density field ( @xcite ; @xcite ) and , second , they move over time as they are accelerated towards the densest regions of the large - scale structure by gravity @xcite .
these two phenomena generally go under the name of `` lagrangian biasing '' and `` lagrangian to eulerian passage '' , respectively . @xcite
dealt with the second problem by assuming that long - wavelength density perturbations evolve according to the spherical top - hat model .
a more sophisticated generalization of the peak - background split has been presented by @xcite who assumed that also the large - scale motion of the density `` peaks '' is fully determined by the long - wavelength component of the density field . since the halo population and the matter feel the same large - scale gravitational potential , their density fluctuations are strongly coupled and their time evolution must be solved simultaneously .
this makes the process of halo biasing non - linear and non - local even if one starts from a linear and local lagrangian biasing scheme ( @xcite ; @xcite ) .
the bispectrum can be used to test this model against the standard eulerian local biasing scheme @xcite . in this paper
, we present a novel and very promising approach to model the clustering of dark matter haloes . adopting the formalism by @xcite combined with a non - local lagrangian biasing scheme for the haloes @xcite
, we simultaneously follow the growth of perturbations in the matter and in the halo distribution over cosmic time .
we present perturbative solutions for the corresponding overdensity and velocity fields and we are able to resum the perturbative series in the limit of large wavenumbers .
moreover , we write down a system of equations for the power spectra @xmath14 and @xmath15 using the time - renormalization - group ( trg ) approach by @xcite and numerically integrate them .
our results are in excellent agreement with the output of a high - resolution n - body simulation , showing an improvement over linear theory , and we are able to predict the matter - halo cross spectrum with a precision within @xmath2 per cent for @xmath16 mpc@xmath4 .
related work has been very recently presented by @xcite who computed the two - point correlation function of linear density peaks and followed its time evolution assuming that peaks move according to the zeldovich approximation . for massive haloes
this results in a scale - dependent bias ( with variations of @xmath17 per cent ) on the scales relevant for baryonic - oscillation studies .
contrary to their approach , we do not deal with a point process but describe large - scale fluctuations in the distribution of dark - matter haloes as perturbations in a continuous fluid . on the other hand ,
we account for the full gravitational motion of the haloes and do not rely on simplified dynamical models as like as the zeldovich approximation .
the structure of our paper is as follows . in section [ mod ]
we present our model for the joint evolution of the matter and halo power spectra .
the initial conditions for our evolutionary equations are discussed in section [ init ] .
the solution of the linearized equations is presented in section [ lin ] where we also quantify the importance of the halo velocity bias .
using a perturbative technique , in section [ analit ] we compute analytic solutions for the propagator of perturbations ( the two - time correlator ) .
we derive 1-loop corrections and , in the limit of large wavenumbers , the fully resummed propagator .
the discussion in sections [ comp1 ] and [ comp2 ] is very technical and the less experienced readers can safely skip it without compromising understanding of the remainder . in section [ num ] , we numerically integrate the full equations for the evolution of halo and matter power spectra in the trg formalism .
we then compare the results against the outcome of a high - resolution n - body simulation .
finally , in section [ con ] we conclude .
the large - scale structure observed today in the universe is believed to be the result of gravitational amplification of primordial fluctuations , caused by the interaction among cold dark matter ( cdm ) particles .
if we denote @xmath9 as the matter density contrast and @xmath18 as the velocity , the eulerian dynamics of a system of such particles , which interact only via gravity , is ruled by a set of three equations ( continuity , euler and poisson ) that in a @xmath19cdm model reads : @xmath20=0\,,\nonumber\\ & & \frac{\partial\,{\bf v}}{\partial\,\tau}+{\cal h}\,{\bf v}\ , + ( { \bf v } \cdot { \bf \nabla } ) { \bf v}= - { \bf \nabla } \phi\,,\nonumber\\ & & \nabla^2 \phi = \frac{3}{2}\,\,{\cal h}^2 \,\omega_m \ , \delta_m \ , , \label{a}\end{aligned}\ ] ] where @xmath21 is the conformal time . if we define the velocity divergence @xmath22 and switch to fourier space , the equations in ( [ a ] ) become : @xmath23 where @xmath24 equations ( [ b ] ) can be written in a compact form if we define a new time variable @xmath25 , being @xmath26 the growth factor at an early epoch , and a doublet @xmath27 ( @xmath28 ) @xmath29 with @xmath30 .
the velocity divergence is scaled such that it has the same dimension of the density contrast and in the linear regime @xmath31 , i.e. @xmath32 . the system is therefore @xmath33 where sum over repeated indices and integration over repeated momenta are understood .
the vertex function @xmath34 ( @xmath35 ) has only three non - vanishing elements @xmath36 and @xmath37 , with @xmath38 the dirac - delta distribution .
all the information about the cosmological model is contained in the matrix @xmath39 that , in the following , will be considered as a constant matrix , approximating @xmath40 .
this ratio is indeed very close to unity for most of the history of the universe . making this approximation ,
one is modifying the behavior of the decaying mode , while the growing one is left unaltered .
it has been shown in @xcite that it affects the matter power spectrum at @xmath1 at a less than percent level up to @xmath41 mpc@xmath4 .
+ the power spectrum , defined by an ensemble average , in this notation is a @xmath42 matrix @xmath43 and the bispectrum , defined by @xmath44 has 8 components . in the following , we will also consider a different - time two - point correlator , defined as @xmath45 which obviously coincides with ( [ h ] ) for @xmath46 .
let us consider a set of dark - matter haloes identified at a given redshift @xmath47 according to some predefined criterion .
the material that forms the haloes can be traced back to its initial location in the linear overdensity field at @xmath48 .
we dub each of these regions as a proto - halo . in other words ,
a proto - halo is the lagrangian region of space that will collapse to form a halo at redshift @xmath47 .
n - body simulations show that nearly all proto - haloes are simply connected @xcite even though this property is not key to our analysis .
let us now follow the evolution of a proto - halo over cosmic time in eulerian space . basically its shape and topology will be distorted ( proto - haloes will first fragment into smaller substructures that will later merge to form the final halo ) and its overall volume will be compressed while its centre of mass will move along a given trajectory determined by the mass density field via gravity .
we focus our analysis onto this motion that connects the lagrangian position of the proto - halo with the eulerian location of the final halo . on scales
much larger than the characteristic size of ( and separation between ) the proto - haloes , the density fluctuations traced by the centre - of - mass trajectories can be described with a continuous overdensity field @xmath49 .
note that while @xmath21 labels conformal time along the trajectories , @xmath47 is just a tag that identifies the halo population . unlike real haloes that undergo merging , by construction proto - haloes always preserve their identity .
their total number is therefore conserved over time and we can write a continuity equation for their number density : @xmath50=0\;. \label{j}\ ] ] here the proto - halo density and velocity fields should be intended as coarse grained on a scale of a few times the mean inter - halo separation ( so as to suppress discreteness effects as proto - haloes are individually separate units ) . strictly speaking , the smoothed velocity field does not obey the euler - poisson system in equation ( [ a ] ) due to the presence of the non - linear term @xmath51 .
in fact , the coarse graining procedure introduces new terms in the fluid equations generated by the degrees of freedom one has integrated out , namely : a velocity dispersion term and a correction to the mean- field gravitational acceleration due to density fluctuations on scales smaller than the smoothing radius ( @xcite ) .
on the other hand , it is reasonable to assume that the large - scale motion of the proto - haloes is generated by density fluctuations with wavelength larger than the characteristic halo size and is not influenced by perturbations with shorter wavelength .
the very same assumption of neglecting the coupling to the small scales is routinely done when one writes equation ( [ a ] ) for the mass density field ( see section 3 in @xcite ) albeit adopting much narrower smoothing kernels than for the haloes . with the same spirit , in what follows , we will ignore the extra terms in the fluid equations generated by the coarse graining procedure .
this is a working hypothesis which makes the problem mathematically treatable and whose accuracy will be tested by comparing our final results against high - resolution numerical simulations .
we therefore write an euler equation for the proto - halo fluid velocities @xmath52 where the gravitational potential is the same as in equation ( 1 ) .
note that if @xmath53 matches @xmath18 in the initial conditions then it will always do . on the contrary ,
any initial velocity bias between proto - halos and matter will be progressively erased by the gravitational acceleration .
thus , given suitable initial conditions for @xmath54 and @xmath53 at @xmath55 ( i.e. a prescription for the lagrangian biasing of proto - haloes ) , we can in principle use equations ( [ j ] ) and ( [ jj ] ) to follow the clustering evolution of the proto - halo population at all times .
we are particularly interested in the solution of the fluid equations at the special time @xmath21 that corresponds to @xmath47 .
in fact this solution has a particular physical meaning as it gives the density and velocity fields of the actual dark - matter haloes . the system ( [ a ] )
is now extended by the inclusion of eq .
( [ j ] ) and eq . ( [ jj ] ) .
we define a quadruplet @xmath27 ( @xmath56 ) @xmath57 in such a way that eq .
( [ e ] ) still holds , but with indices running from 1 to 4 .
there are three more non - vanishing elements of the vertex @xmath58 and @xmath59 , and the @xmath60 @xmath61 matrix is @xmath62 from the definitions ( [ h ] ) and ( [ i ] ) , with @xmath56 , we get a @xmath63 matrix for power spectrum ; in the following , we will focus on the matter power spectrum @xmath64 and the matter - halo cross spectrum @xmath65 .
in the previous section , we have presented a model that describes the non - linear evolution of the matter and halo density fields . given suitable initial conditions , the formal equations we have introduced can be integrated numerically so that to compute the perturbative propagators and the trg - evolved power spectra .
the choice of the initial conditions therefore plays a very important role in our theory and will be the subject of this section . to gain insight into the properties of proto - haloes ( and , later , to test our results at @xmath1 ) , we use one high - resolution n - body simulation extracted from the suite presented by pillepich et al .
this contains @xmath66 dark - matter particles within a periodic cubic box with a side of @xmath67 mpc and follows the formation of structure in a @xmath19cdm model with gaussian initial conditions and cosmological parameters : @xmath68 , @xmath69 , @xmath70 , @xmath71 , @xmath72 and @xmath73 .
we identify dark - matter haloes at @xmath1 using the friends - of - friends algorithm with a linking length equal to 0.2 times the mean interparticle distance .
we only consider haloes containing more than 100 particles ( i.e. with mass @xmath74 ) and we split them into four mass bins to keep track of their different clustering properties .
the corresponding mass ranges and the total number of haloes in each bin are given in table [ tab:1 ] , along with an estimate of the highest wavevector up to which the fluid approximation for haloes holds .
this value is determined by the number of haloes we require to be in a volume element to consider them as a fluid , and we set this number to 30 . on smaller scales ,
our assumption breaks down , therefore we will look at results in the specified range , that , of course , decreases as the mass of the haloes increases . in the plots that will be shown in section [ num ] the limit to which we can trust our model will be represented by vertical black dotted lines .
proto - haloes are identified by tracing the positions of the particles forming a halo at @xmath1 back to the linear density field .
the centre of mass of each proto - halo is used as a proxy for its spatial location .
similarly , the mass weighted linear velocity gives the proto - halo velocity .
halo and proto - halo density and momentum fields are computed with the cloud - in - cell grid assignement using a @xmath75 mesh .
velocity fields are obtained by taking the ratio of the momentum and density distributions ( preventively smoothed to preclude the existence of empty cells ) as shown in @xcite .
power spectra have been computed using fft . in order to avoid uncertain shot - noise corrections for the haloes
, we only consider their cross spectra with the matter density field . concerning the matter density ,
the initial conditions are given by linear theory which directly follows from the adopted cosmological model ( transfer function ) and the statistics of primordial perturbations ( spectral index , gaussianity ) .
on the other hand , for the dark - matter halos , we can follow two different approaches : ( i ) extract the relevant information directly from the simulation or ( ii ) use a model for the lagrangian bias of the halos .
the latter option offers a number of advantages .
first , it allows us to make general predictions independently of the simulation specifics .
second , it allows us to include halo bispectra in our formalism ( while it would be extremely demanding and time consuming to compute all possible triangular configurations from the simulation ) . for these reasons we will present below a model for the bias of the proto - haloes .
note , however , that any lack of accuracy of the adopted lagrangian biasing scheme will propagate through the time evolution of our model and contribute to the imprecision of its final results .
therefore , in order to test the accuracy of our evolutionary equations alone , we will also extract initial conditions directly from the simulation and compare the corresponding outcome of the evolution model with the statistics of the simulated haloes at @xmath1 .
let us consider the overdensity of proto - haloes in lagrangian space @xmath76 and the corresponding mass - density fluctuation @xmath77 .
we assume that their fourier transforms are linked by the expression : @xmath78 which corresponds to a non - local relation in real space .
this form was first proposed by @xcite and describes the clustering of linear density peaks @xcite . in this case , the initial conditions for @xmath79 and @xmath65 are : @xmath80 where the exponential functions accounts for the finite size of the density peaks corresponding to a given halo mass .
we find that the expression above accurately describes the cross - spectrum of proto - haloes and matter in our n - body simulation when @xmath6 , @xmath81 and @xmath82 are treated as fitting parameters ( see @xcite for further details ) .
, provides a worse fit to simulation results . in this case ,
the model @xmath65 lacks power for all but the smallest wavenumbers . ]
this is not surprising as @xcite have shown that most proto - haloes include a density peak of the corresponding mass scale within their lagrangian volume . in table
[ tab:1 ] we quote , for each halo - mass bin , the parameters @xmath6 , @xmath81 and @xmath82 that best fit the simulation data using eq . ( [ x ] ) where the linear matter power spectrum @xmath64 is computed using the camb online tool @xcite .
whereas gaussianity is a good approximation for the linear matter distribution , fluctuations in the halo counts are non - gaussian even in the initial conditions .
we can quantify their level of non - gaussianity in terms of their auto and cross bispectra that , using equation ( [ w ] ) , can all be reduced to one of the following forms : @xmath83 where the matter bispectrum @xmath84 is computed using the tree - level expression of the standard perturbation theory , @xmath85\nonumber\\ & & p_{11}(k_1,z_{in})p_{11}(k_2,z_{in})+{\rm cycl.}\,,\end{aligned}\ ] ] with @xmath86 .
+ in order to solve our evolutionary equations , we need to know also the linear velocity field of proto - haloes . in principle proto - haloes might not move with the same velocity as matter at the same location ( at the very least they should match the mass velocity smoothed on the lagrangian halo size ) .
we model this effect assuming that proto - haloes are indeed related to linear density peaks which , as discussed above , gives a good description of their clustering properties .
in particular we follow @xcite who proposed a model for the peak velocities , which in fourier space assumes the form _ h()=(1-k^2 ) e^-k^2r^2/2 _ m()b_v()_m ( ) , [ vbias ] with @xmath87 the scale - dependent `` velocity bias '' and @xmath88 ( @xmath89 ) being the spectral moments of the matter power spectrum defined as _ n^2 = _ 0^ kk^2(n+1)p_11(k)e^-k^2r^2 . in the equations above a gaussian smoothing window of size @xmath82 has been adopted to coarse grain the matter density and identify the peaks .
once the proto - halo mass is linked to @xmath82 through @xmath90 ( with @xmath91 the mean comoving density of matter ) , this model has no free parameters , since the spectral moments are completely defined by @xmath92 .
it follows that the initial conditions for @xmath93 and @xmath94 are @xmath95 and these expressions are in very good agreement with the spectra computed from the n - body simulation ( @xcite ) .
the corresponding values of @xmath96 are listed in table [ tab:1 ] as a function of halo mass .
note that the velocity bias becomes more and more important on large scales with increasing the halo mass .
[ cols="^,^,^,^,^,^,^,^,^ " , ]
the lowest order approximation to the perturbation equation ( [ e ] ) consists in setting @xmath97 . in this limit
, the evolution of the field from the initial time @xmath98 to a generic @xmath99 is given by _
a(;)=g_ab()_b(;0 ) , [ new3 ] where @xmath100 is the _ linear propagator _ , defined by the equation ( _ ab _
+ _ ab ) g_bc ( ) = _ ac _ d ( ) , [ new4 ] with @xmath101 the kronecker delta . solving eq .
( [ new4 ] ) with causal boundary conditions ( @xmath102 for @xmath103 , see e.g. @xcite ) one gets @xmath104\,\theta(\eta)\ , , \label{new5}\end{aligned}\ ] ] with @xmath105 heavyside s step function .
notice that @xmath106 as @xmath107 .
the first and second contributions represent the standard growing and decaying modes , respectively @xcite .
the third and fourth contributions represent two new modes , decaying respectively as @xmath108 and @xmath109 compared with the growing one . to understand their physical effect
we notice that an initial condition of the form _
a(;0)= ( c ( ) + ( ) + _ h ( ) + _ v ( ) + ) , [ ic ] evolves ( using eq .
( [ new3 ] ) ) into @xmath110 given by ( c + + + 2 e^-3/2(-_v)+ e^-(-3+_h+2_v ) + -e^-3/2(-_v ) + ) , [ uff ] both the halo density and velocity fields relax to the matter ones as @xmath111 ( but at a different pace ) . also note that in the absence of an initial density bias ( i.e. @xmath112 ) but in the presence of an initial velocity bias ( i.e. @xmath113 ) , the linear dynamics quickly generates a transient density bias that vanishes at late times as @xmath114 . the initial power spectrum at @xmath98 , corresponding to the field configuration in eq . ( [ ic ] ) , is @xmath115 , and it evolves forward in time as @xmath116 it is interesting to assess the role of the velocity bias in the linear solution previously discussed .
assuming @xmath117 at all times , the linear propagator for the first three components @xmath118 with @xmath119 becomes @xmath120\,\theta(\eta)\ , , \label{n}\end{aligned}\ ] ] and the third component of ( [ uff ] ) reduces to _ 3(;)=()+e^-(_h()- ( ) ) .
this expresses the well known
_ linear debiasing _ between the halo and matter fields at late times , derived by @xcite for tracers that do not undergo merging and move solely under the influence of gravity .
the corresponding halo - matter cross spectrum is p_l,13=p^0_11 + e^-(p^0_13 - p^0_11)p_l,13^(3 ) , while keeping @xmath121 one gets p_l,13=p_l,13^(3 ) + 2 ( p^0_11-p^0_14 ) ( e^-3 /2 - e^- ) .
[ vb ] in figure [ fig0 ] we compare @xmath122 against @xmath123 extracted from the n - body simulation , for the different mass bins .
while the spectra agree well on very large scales ( @xmath124 ) , they progressively depart for smaller scales .
this is in line with the model introduced in section 3 .
note that the last term in eq .
( [ vb ] ) vanishes in the initial conditions , reaches a minimum for @xmath125 and it is suppressed at late times .
we quantify its amplitude at @xmath1 in figure [ fig2 ] , where we plot the ratio @xmath126 which ranges between @xmath127 per cent , depending on halo mass and scale .
this suggests that the effect of the velocity bias on the halo - matter cross spectrum is small for low redshifts .
in this section we deal with the non - linear evolution of the matter - halo system .
we first compute 1-loop corrections for the propagator and then perform the corresponding large - k resummation to all perturbative orders thus extending the results presented by @xcite for the matter density field .
finally , we compute non - linear power spectra using the trg approach . for simplicity we only consider the case with no velocity bias , which we have demonstrated to be accurate ( at least in the linear regime ) at low redshifts , where current observations are available .
the @xmath128 @xmath61 matrix for this case is @xmath129 the 1-loop correction to the linear propagator @xcite is given by @xmath130 ( see appendix a for its explicit expressions ) .
the 1-loop correction to the two - point correlator ( [ 2pc ] ) is given by the sum of two contributions p_ab^i(k;_a,_b)+p_ab^ii(k;_a,_b ) , which are also known in the literature as @xmath65 " and @xmath131 " , respectively . and
@xmath131 of our notation ! ]
they are given by p_ab^i(k;_a,_b ) & = & g_ac(k;_a ) g_bd(k;_b ) p_cd^0(k ) + & & + ( ab ) , p_ab^ii(k;_a,_b ) & = & _ 0^_a ds_a _ 0^_b ds_b _ cd(k;s_a , s_b ) + & & g_ac(k;_a - s_a ) g_bd(k;_b - s_b ) , with _ cd(k;s_a , s_b ) & = & 2 e^s_a+s_b d^3q + & & _ cei(*k*,*-q*,*q - k*)p_l , ef(q , s_a , s_b ) + & & p_l , ih(|*k*-*q*|,s_a , s_b)_dfh(*-k*,*q*,*k - q * ) .
+ [ phi ] the explicit expressions for @xmath132 are given in appendix a. in the large-@xmath133 limit , the 1-loop correction for the propagator , eq .
( [ r ] ) , grows as @xmath134 , and eventually dominates over the ( @xmath133-independent ) linear propagator .
taking into account higher orders , the situation gets even worse .
the 2-loop correction grows as @xmath135 , the 3-loop as @xmath136 , and so on .
this a manifestation of the perturbative expansion breakdown in cosmological pt , which appears not only in the computation of the propagator , but also of the power spectrum , the bispectrum , and so on .
however , for the case of the propagator , it was shown in @xcite that the leading order corrections in the large @xmath133 and large @xmath99 limit can be resummed at all orders in perturbation theory , giving a well - behaved propagator .
+ the propagator @xmath137 connects the initial correlators with the cross - correlations between final and initial field configurations , p_ab(k ; , 0 ) = g_ac(k ; ) p^0_cb(k ) + p_ab^ng(k;,0 ) , [ ct ] where the last term at the rhs comes from the initial non - gaussianity of the matter and halo fields . at leading order
, it is given by ( see appendix [ ang ] ) @xmath138 where @xmath139 is the initial bispectrum at @xmath98 ( @xmath140 ) ( see also @xcite ) .
in section [ propres ] we will use eq .
( [ ct ] ) to assess the validity of different approximation schemes for the propagator . in the large-@xmath133 limit
, @xmath141 decays as @xmath142 with @xmath143 therefore , at least in the case of the propagator , the bad ultraviolet behavior is just an artifact of the perturbative expansion , which , at any finite order , completely misses the nice and physically motivated gaussian decay of eq .
( [ s ] ) ( see @xcite and @xcite for a detailed discussion ) .
although the result ( [ s ] ) was obtained for the @xmath144 propagator of the matter density - velocity system , it holds , taking into account proper modifications , also when halos are included , _
i.e. _ for the @xmath145 propagator considered in this section . as in @xcite , we obtain an improved propagator interpolating between the 1-loop result ( eq . [ r ] ) at low @xmath133 and the gaussian decay ( [ s ] ) ( with a modified pre - factor ) at high @xmath133 .
the details of the derivation and the relevant formulae are given in appendix b. unlike the propagator , the power spectrum can not be resummed analytically at large k. different semi - analytical procedures ( @xcite , @xcite , @xcite ) have been proposed to compute it in the mildly non - linear regime . in this paper
we will consider the trg technique introduced in @xcite .
starting from eq .
( [ e ] ) , a hierarchy of differential equations for the power spectrum , the bispectrum and higher order correlations is obtained .
we choose to truncate it at the level of the trispectrum @xmath146 , so that the equations for @xmath147 and @xmath139 form a closed system @xmath148\,,\nonumber\\ & & \nonumber\\ & & \partial_\eta\,b_{abc}({\bf k},\,{\bf -q},\,{\bf q - k};\,\eta ) = \nonumber\\ & & \qquad\quad - \omega_{ad } ( \eta)b_{dbc}({\bf k},\,{\bf -q},\,{\bf q - k};\,\eta)\nonumber\\ & & \qquad\quad- \omega_{bd } ( \eta)b_{adc}({\bf k},\,{\bf -q},\,{\bf q - k};\,\eta)\nonumber\\ & & \qquad\quad - \omega_{cd } ( \eta)b_{abd}({\bf k},\,{\bf -q},\,{\bf q - k};\,\eta)\nonumber\\ & & \qquad\quad + 2 e^\eta \left [ \gamma_{ade}({\bf k},\,{\bf -q},\,{\bf q - k } ) p_{db}(q;\,\eta)p_{ec}(|{\bf k}-{\bf q}|;\,\eta)\right.\nonumber\\ & & \qquad\quad + \gamma_{bde}({\bf -q},\,{\bf q - k},\,{\bf k } ) p_{dc}(|{\bf k}-{\bf q}|;\,\eta)p_{ea}(k;\,\eta)\nonumber\\ & & \qquad\quad + \left .
\gamma_{cde}({\bf q - k},\,{\bf k},\,{\bf -q } ) p_{da}(k;\,\eta)p_{eb}(q;\,\eta)\right]\ , , \label{m}\end{aligned}\ ] ] which integrated gives the power spectra at any redshift and for any momentum scale .
the system ( [ m ] ) consists of coupled differential equations which are solved numerically , starting from given initial conditions , i.e. @xmath149 and @xmath150 .
+ from eq .
( [ l ] ) , we can observe that @xmath151 and @xmath152 are zero , which means that the evolution of @xmath9 and @xmath153 is not modified , with respect to the original trg formulation , by the presence of @xmath154 , as it is expected .
in order to assess the validity of our analytical approach , we compare our results for the resummed propagator against the simulation ; to this end , we consider the relation in eq .
( [ ct ] ) .
in particular we choose the indices @xmath155 , so that we can check the components related to the haloes that were not present in the original formalism by @xcite and , at the same time , we do not have to deal with the shot - noise problem .
we extract the cross spectra from the simulation and compare them against those obtained both using linear theory propagators @xmath156 and resummed propagators @xmath157 ; the result is shown in figure [ fig9 ] .
we note that the linear model severely overpredicts the two - time cross spectrum for @xmath158 .
it is evident that the resummed theory improves considerably upon the linear one , and agrees with the simulation within 10 per cent accuracy up to the scale where the fluid approximation holds .
we include in @xmath159 the effect of the initial non - gaussianity of the halo field via its initial bispectra , computed as in eq .
( [ y ] ) .
it turns out that the components giving a non - vanishing effect are of the @xmath160 type .
their contribution is suppressed by a @xmath161 factor with respect to that of an hypothetical primordial non - gaussianity in the matter field ( which we do not consider here ) .
therefore , the effect is of modest entity but , nevertheless , it improves the agreement with the simulation with respect to the case in which it is neglected .
the vertical black dotted lines represent the limit up to which we expect our fluid approximation for the haloes to work .
[ fig9 ] the trg equations presented above are integrated numerically starting from the initial conditions discussed in section [ init ] . as a first step , we set all the initial bispectra to zero in figure [ fig3 ] we show a comparison between the halo - matter cross spectra extracted from the n - body simulation and the results of trg , one - loop and linear theory . in the first three bins , corresponding to lower halo masses , linear theory overpredicts the power on mildly non - linear scales ;
note that this departure arises on larger scales compared to the matter auto spectrum ( not shown in the figure ) .
the overprediction of linear theory is cured by the one - loop power spectrum only on very large scale , while the trg manages to correct it up to a smaller scale , before starting to fail .
the fourth bin , though , displays a different behaviour : linear theory lacks of power on small scales , and neither the 1-loop correction nor the trg are much more accurate .
this might originate from the fact that very massive haloes are large and rare in the initial conditions , therefore less suited for the fluid approximation .
moreover , they also display the strongest velocity bias , which we are neglecting in our current non - linear treatment .
a more quantitative analysis is presented in figure [ fig5 ] , where we plot the ratios between the spectra from the simulation and the theoretical results .
the trg gives a cross spectrum within @xmath162 accuracy at least up to @xmath163 ( barring bin 4 ) , while linear theory does so up to @xmath164 .
+ as previously pointed out , however , the halo - density field is not gaussian at @xmath165 .
we use expressions ( [ y ] ) to account for initial bispectra in the trg method ( they are not present in the linear theory and at 1-loop level ) .
figure [ fig5 ] illustrates that this indeed produces a slight improvement in the agreement with the simulation .
the correction resulting from the introduction of the initial bispectra turns out to be quite small .
the reason is that their contribution to the final cross spectrum @xmath65 is suppressed . to understand why
, we can use perturbation theory ; first , let us investigate the case of @xmath64 .
in the one - loop computation , the initial @xmath64 contributes to the final one with a weight " of @xmath166 , as it also happens in linear theory .
the initial matter bispectrum @xmath167 , instead , has a weight of @xmath168 , so it is suppressed by a factor of @xmath109 .
factor is replaced by the bispectrum . ]
if we now consider the haloes , we have showed in ( [ n ] ) that there is a new decaying mode , responsible for the linear debiasing .
this new mode , that involves only the halo field , carries an extra @xmath109 suppression factor .
we can now rank the contributions to @xmath65 according to their relevance : 1 .
@xmath64 2 .
@xmath65 , @xmath169 3 .
@xmath79 , @xmath170 4 .
@xmath171 5 .
@xmath172 .
each item is suppressed by a factor of @xmath109 with respect to the previous one .
we can see that only @xmath167 has some relevance , while the other terms are highly suppressed .
even if the reasoning was based on perturbation theory , it is valid also for trg , at a qualitative level .
incidentally , the fact that @xmath64 is the most relevant contribution is another evidence of debiasing .
+ we can now address the effect of the truncation we perform in the trg , namely considering the trispectrum @xmath173 .
first of all , the matter trispectrum @xmath174 can be neglected in the range of scales under consideration , as one can conclude from the comparison between trg and simulations in @xcite . the contribution from initial mixed ( i.e. matter - halo ) or pure halo trispectra
is furtherly suppressed with respect to @xmath174 by extra @xmath109 factors , for the same reason as above .
however , the trispectrum has its own time evolution as well , and one might argue that it becomes more relevant for @xmath175 ; this seems to be not the case , because enlarging the trg truncation scheme by including the running of the trispectrum gives a contribution to the power spectrum which is at least of two - loop order and it is certainly subdominant in the scales we are considering .
+ from section 5 onward , we neglected any velocity bias , since this approximation proved to be accurate enough at @xmath1 ( in linear theory ) . as a further check , it is interesting to observe from figure [ fig4 ] that the trg is able to give a better prediction than full linear theory in eq .
( [ vb ] ) , even neglecting the velocity bias that the linear theory accounts for .
+ we can now look at the model predictions for the halo bias , defined as the ratio @xmath176 .
this quantity is plotted as a function of wavenumber in figure [ fig6 ] .
while the linear - theory bias always increases with scale , irrespectively of halo mass , the trg result closely follows the scale dependence of @xmath11 seen in the simulation for the first two bins .
it also gives a nearly constant bias for the third bin , as the simulation does , even though with a slightly lower value .
however , the linear model performs better in the last bin where the bias in the simulation increases with @xmath133 .
it is also interesting to look at the results displayed in a different way ; we can investigate the evolution of the cross spectrum from the initial conditions to today and the evolution of the bias as well . in figures
[ fig7 ] and [ fig8 ] we plot , respectively , @xmath177 again , our model is able to match accurately the trend of the simulation and to improve upon linear theory , excluding bin 4 .
a key feature of the linear solution in eq . ( [ uff ] ) is the debiasing between halo and matter distributions with time .
it is worth noting from figure [ fig8 ] that this effect is stronger for high - mass haloes , but constant on all the scales , while for low - mass haloes it is weaker on large scales and presents a strong @xmath133-dependence .
we have presented a novel approach to modeling the clustering of dark - matter haloes on mildly non - linear scales .
this follows the motion of the regions that will collapse to form haloes ( that we dub proto - haloes ) .
since the number of proto - haloes is conserved over time , for sufficiently large scales ( @xmath178 ) , we can write a set of fluid equations that govern their evolution under the effect of gravity , which couples perturbations in the halo and matter density fields . we provide analytical solutions for the linearized equations and 1-loop perturbative corrections for the halo and matter power spectra . for the propagator , quantifying the memory of the density and velocity fields to their initial conditions , we also perform a resummation of perturbative corrections .
finally , for the power spectrum we compute the non - linear evolution using a semi - analytical procedure , namely an extension of the time renormalization group .
the initial conditions for our evolutionary equations are specified adopting a lagrangian bias model , originally developed to describe the clustering and motion of linear density peaks .
we fix the parameters of the model so that to reproduce the distribution of proto - haloes in a high - resolution n - body simulation at @xmath179 .
we use the same simulation to test the predictions of our model at @xmath1 .
our main results can be summarized as follows : * independently of the initial conditions , in the linear solution the halo density and velocity fields asymptotically match the corresponding matter fields at late times .
this debiasing develops at a different rate for the density and the velocity , being faster for the latter .
* even if there is no initial density bias , the presence of a velocity bias generates a transient density bias that vanishes at late times . *
neglecting any initial velocity bias alters the linear predictions for the halo - matter cross spectrum at redshift @xmath1 only by less than 3 per cent , for @xmath180 .
this provides us with the motivation to ignore the velocity bias in the non - linear analysis .
* unlike its linear counterpart , the resummed propagator is in good agreement with the n - body simulation , independently of halo mass . *
the halo - matter cross spectrum predicted by the trg is accurate to @xmath2 per cent up to @xmath181 for a broad range of halo masses .
this does not hold for very massive haloes ( @xmath182 ) , that have low number density and high initial velocity bias , for which discreteness effects are more important . * the trg result improves upon both linear theory and 1-loop corrections . its performance is slightly enhanced accounting for the initial non - gaussianity of the halo distribution . * for low halo masses
our model accurately describes the scale - dependent bias measured in the simulation at @xmath1 .
ae was supported through the sfb - transregio 33 the dark universe " by the deutsche forschungsgemeinschaft ( dfg ) .
+ ae and sk acknowledge bcgs support for part of the work . +
cp thanks the participants to the workshop
modern cosmology : early universe , cmb and lss " held at the centro de ciencias de benasque pedro pascual for stimulating discussions .
part of this work has been developed there in august 2010 .
99 bardeen j.m . ,
bond j.r . , kaiser n. , szalay a.s . , 1986 , apj , 304 , 15 bartolo n. , beltran almeida j.p . , matarrese s. , pietroni m. , riotto a. , 2010 , jcap 1003 , 011 bernardeau f. , crocce m. , sefusatti e. , 2010 , phys . rev .
d , 82 , 083507 buchert t. , dominguez a. , 2005 , a & a , 438 , 443 catelan p. , lucchin f. , matarrese s. , porciani c. , 1998 , mnras , 297 , 692 catelan p. , porciani c. , kamionkowski m. , 2000 , mnras , 318 , l39 cole s. , kaiser n. , 1989 , mnras , 237 , 1127 cole s. et al . , 2005 , mnras , 362 , 505 coles p. , 1993 ,
mnras , 262 , 1065 cooray a. , sheth r.k . , 2002 , phr , 372 , 1 crocce m. , scoccimarro r. , 2006 , phys .
d , 73 , 063519 crocce m. , scoccimarro r. , 2006 , phys . rev .
d , 73 , 063520 desjacques v. , 2006 , phys .
d , 78 , 103503 desjacques v. , sheth r.k .
, 2010 , phys .
d , 81 , 023526 desjacques v. , crocce m. , scoccimarro r. , sheth r.k .
, 2010 , phys .
d , 82 , 103529 eisenstein d.j .
, hu w. , 1998 , apj , 496 , 605 elia a. , ludlow a. , porciani c. , _ in preparation _ fry j.n .
, 1996 , apj , 461 , l65 fry j.n . , gaztaaga e. , 1993 , apj , 413 , 447 guzik j. , bernstein g. , smith r.e . , 2007 mnras , 375 , 1329 huff e. , schulz a. e. , white m. , schlegel d.j . , warren , m.s . , 2007 , aph , 26 , 351 kaiser n. , 1984 , apj , 284 , l9 lewis a. , challinor a. , lasenby a. , 2000 , apj , 538 , 473 ludlow a. d. , porciani c. , 2010 , arxiv 1011.2493 manera m. , gaztaaga e. , 2009 , arxiv 0912.0446 manera m. , sheth r.k .
, scoccimarro r. , 2010 , mnras , 402 , 589 mann , r. g. , peacock , j. a. , heavens , a. f. , 1998 , mnras , 293 , 209 matarrese s. , pietroni m. , 2007 , jcap , 6 , 26 matsubara t. , 1999 , apj , 525 , 543 matsubara t. , 2008 , phys . rev .
d , 78 , 3519 mcdonald p. , 2007
d , 75 , 043514 mo h.j . ,
white s.d.m . , 1996 , mnras , 282 , 347 montesano f. , sanchez a.g .
, phleps s. , 2010 , arxiv 1007.0755 norberg p. et al .
, 2002 , mnras , 332 , 827 pietroni m. , 2008 , jcap , 10 , 36 pillepich a. , porciani c. , hahn o. , 2010 , mnras , 402 , 191 porciani c. , catelan p. , lacey c. , 1999 , apj , 513 , l99 porciani c. , dekel a. , hoffman y. , 2002 , mnras , 332 , 339 schulz a. e. , white , m. , 2006 , aph , 25 , 172 scoccimarro r. , 2004 , phys .
d , 70 , 083007 seljak u. , 2001 , mnras , 325 , 1359 seo h. , eisenstein d.j .
, 2005 , apj , 633 , 575 smith r.e .
, scoccimarro r. , sheth r.k .
, 2007 , phys .
d , 75 , 063512 sugiyama n. , 1995 , apj , 100 , 281 taruya a. , hiramatsu t. , 2008 , apj , 674 , 617 zehavi i. et al . , 2004 , apj , 608 , 16
linear power spectra : @xmath183 first we report the integrands of the one - loop corrections of the components we are interested in . the linear power spectrum can be split as p_l , c_i , c_j(q_i;s_i , s_j ) = p^0(q_i ) u_c_i , c_j + p_c_i , c_j(q_i;s_i , s_j ) , [ app1 ] where u_c_i , c_j = 1 c_i , c_j , [ app2 ] while p_c_i , c_j(q_i;s_i , s_j ) 0 c_i c_j=3 .
[ bw ] looking at eq .
( [ r ] ) we can consistently split the one - loop correction to the propagator as @xmath184 and we denote them as @xmath185 , so that @xmath186 , for @xmath187 .
@xmath188}{840 k^3 q^3}\,,\nonumber\\ \delta^{(1 ) } g_{31}&=&\delta g_{11}\,,\nonumber\\ \delta^{(1 ) } g_{32}&=&\frac{2}{3 } \delta g_{11}\,,\nonumber\\ \delta^{(1 ) } g_{33}&=&-\frac{2}{3 } e^{-\eta } ( e^{\eta } -1)^2 k^2 \pi p^0_{11}(q)\,,\nonumber\\ \delta^{(2 ) } g_{31}&=&\frac{\pi e^{\eta } \left(p^0_{13}(q)-p^0_{11}(q)\right ) \left(4 \left(9 k^5 q+24 k^3 q^3 - 9 k q^5\right)-18 \left(k^2-q^2\right)^3 \log \left(\left|\frac{k+q}{k - q}\right|\right)\right)}{280 k^3 q}\,,\nonumber\\ \delta^{(2 ) } g_{32}&=&\frac{2}{3 } \delta^{(2 ) } g_{31}\,,\nonumber\\ \delta^{(2 ) } g_{33}&= & 0\end{aligned}\ ] ] from eq .
( [ phi ] ) , one can see that the one - loop corrections require integrals over @xmath189 , @xmath190 and @xmath191 ; the integration over @xmath192 and over @xmath193 can not be performed analytically .
the following expressions are therefore integral kernels , denoted by @xmath194 ( the factor of @xmath195 is already taken into account ) . if @xmath196 and @xmath197 , @xmath198 ^ 2 p^0_{11}\left(|k - q|\right)}{49 \left(k^2 - 2 k \mu q+q^2\right)^2}\,,\nonumber\\\end{aligned}\ ] ] _ q p^ii_13&= &
. @xmath199 + 7 q p^0_{33}(q ) ( k-\mu q)\right]\nonumber\\ & & + k^2 p^0_{11}(q ) p^0_{11}\left(|k - q|\right ) \left[q
\left(\left(10 \mu ^2 - 3\right ) e^{\eta } -14 \mu ^2 + 7\right)-7 k \mu \left(e^{\eta } -1\right)\right]^2\nonumber\\ & & + 49 \mu ^2 p^0_{11}(q ) \left(k^2 - 2 k \mu q+q^2\right)^2 p^0_{33}\left(|k - q|\right)\nonumber\\ & & \left.+14 k \mu
p^0_{11}(q ) \left(k^2 - 2 k
\mu q+q^2\right ) \left[7 k \mu \left(e^{\eta } -1\right)-q \left(\left(10 \mu ^2 - 3\right ) e^{\eta } -14 \mu ^2 + 7\right)\right ] p^0_{13}\left(|k - q|\right)\right\}\,.\end{aligned}\ ] ]
as it was discussed in @xcite , the leading contribution in @xmath200 to the propagator at a generic @xmath201-loop order contains a chain of propagators and vertices of the form g_ab_1(-s_1 ) _
b_1c_1a_1(,-_1,_1- ) g_a_1b_2(s_1-s_2 ) g_a_2n-1b_2n(s_2n-1-s_2n ) _
b_2nc_2na_2n(+_2n,-_2n,- ) g_a_2nb(s_2n ) . [ chain ]
the @xmath202 indices have to be contracted in all possible pairings , by inserting @xmath201 linear power spectra , each of the form _
d(_i+_j ) p_l , c_i , c_j(q_i;s_i , s_j ) .
the @xmath201-loop contribution is obtained by multiplying by @xmath203 and by the appropriate combinatoric factor , and then by integrating over _ i=1 ^ 2n ( _ 0^ds_i d^3 q_i ) .
recall eqs .
( [ app1]-[bw ] ) : if all the insertions are of the @xmath204 type then the resummation goes exactly as in the standard case .
indeed u_c_1,c_j g_a b_1(-s_1 ) _
b_1c_1a_1(,-_1,_1- ) g_a_1b_2(s_1-s_2 ) u_c_j g_ab_2(-s_2 ) , in the @xmath205 limit . besides the explicit form for the vertices ( see section 1 )
, we have used the composition property of the propagators g_a b(-s_1 ) g_b c(s_1-s_2)= g_a c(-s_2 ) , and have defined @xmath206 . since each vertex , contracted by a @xmath207 vector becomes proportional to a delta function in its first and third index , the chain of propagators composes up to a single propagator @xmath208 and the time integral
can be easily performed .
the momentum integrals factorize into @xmath201 integrals of the type - d^3 q p^0(q ) = - k^2 ^2 , where @xmath209 has been defined in ( [ t ] ) . using the appropriate combinatoric factors ,
the leading contribution to the propagator at @xmath201-loop , in the large momentum limit , when all the power spectrum insertions are of the @xmath210 type is ^n g_ab ( ) , [ nord ] which resums to @xmath211 $ ] . as for the insertions including the @xmath212 contribution to the linear power spectrum ,
the important point to realize is that , due to ( [ bw ] ) and to the structure of the linear propagator and the vertices , a chain like ( [ chain ] ) can be contracted by _ at most one _ @xmath212 power spectrum , the remaining ones being of the type @xmath204 .
moreover , these insertions only contribute to @xmath213 and @xmath214 .
therefore , at @xmath201-th order , we have only two types of contributions : those with all @xmath204 insertions , giving ( [ nord ] ) , and those with one @xmath215 insertions and @xmath216 @xmath204 ones , which can also be resummed in the large @xmath133 limit . as a consequence ,
the complete resummed propagator in the large momentum limit is g_ab(k ; ) = ( g_ab ( ) + _
a3 f_b ^(2 ) g_31(k ; ) ) , with @xmath217 . in order to have an expression for the propagator valid at any k
, one can proceed as in ( @xcite ) and interpolate between the large @xmath133 limit above and the 1-loop result at low k. this can be done , for instance for @xmath218 , starting from its 1-loop expression g_11()+g_11(k ; ) g_11()(1 + g_11(k ; ) ) , where the above approximation is exact in the large @xmath99 limit . since g_11(k ; ) -k^2 ^2 , k , the required interpolation is given by g_11(k;)=g_11 ( ) .
proceeding analogously for the other components , we have : g_12(k;)&=&g_12 ( ) , + g_22(k;)&=&g_22 ( ) , + g_21(k;)&=&g_21 ( ) , + g_31(k;)&=&(g_31()+^(2 ) g_31(k ; ) ) , + g_32(k;)&=&(g_32 ( ) + ^(2 ) g_31(k ; ) ) , + g_33(k;)&=&g_33 ( ) , where we have used the identities g_12(k ; ) & = & g_11(k ; ) , + g_21(k ; ) & = & g_22(k ; ) , + ^(1 ) g_31(k ; ) & = & g_11(k ; ) + ^(1 ) g_32(k ; ) & = & ^(1 ) g_31(k ; ) + ^(2 ) g_32(k;)&=&^(2 ) g_31(k ; ) .
while at @xmath219 the matter field can be considered gaussian with high accuracy , the same not necessarily holds for haloes .
an initial non - gaussianity for the equal - time power spectrum can be easily incorporated in the trg formalism as discussed in @xcite , and has been done in sect .
6.2 . in order to obtain the effect of an initial non - gaussianity on the cross - correlator at different times considered in eq .
( [ ct ] ) , it suffices to compute the @xmath220 correction to the linear evolution of the field , by inserting eq .
( [ new3 ] ) in ( [ e ] ) , to get _
a^(1 ) ( ; ) = _ 0^ds g_ac(-s)e^s d^3 q _
cde(,-,- ) g_df(s ) _
f(;0 ) g_eg(s ) _ g(-;0 ) . the cross - correlator _
a^(1 ) ( ; ) _ b ( ^;0 ) , then includes the non - gaussian expectation value _ f(;0 ) _
g(-;0 ) _ b ( ^;0)= _ d(+^ ) b_fgb(,-,- ) , and therefore gives eq . ( [ eng ] ) . | we address the issue of the cosmological bias between matter and galaxy distributions , looking at dark - matter haloes as a first step to characterize galaxy clustering . starting from the linear density field at high redshift , we follow the centre of mass trajectory of the material that will form each halo at late times ( proto - halo ) .
we adopt a fluid - like description for the evolution of perturbations in the proto - halo distribution , which is coupled to the matter density field via gravity .
we present analytical solutions for the density and velocity fields , in the context of renormalized perturbation theory .
we start from the linear solution , then compute one - loop corrections for the propagator and the power spectrum .
finally we analytically resum the propagator and we use a suitable extension of the time - renormalization - group method @xcite to resum the power spectrum . for halo masses
@xmath0 our results at @xmath1 are in good agreement with n - body simulations .
our model is able to predict the halo - matter cross spectrum with an accuracy of @xmath2 per cent up to @xmath3 mpc@xmath4 approaching the requirements of future galaxy redshift surveys .
[ firstpage ] cosmology : theory , large - scale structure of universe galaxies : haloes methods : analytical , n - body simulations . |
in order to be manipulated and transmitted information should be encoded into some degree of freedom of a physical system . ultimately , this means that the input alphabet should correspond to the spectrum of some observable , _ i.e. _ that information is transmitted using _
quantum signals_. at the end of the channel , to retrieve this kind of quantum information , one should measure the corresponding observable . as a matter of fact
, the measurement process unavoidably introduces some disturbance , and may even destroys the signal , as it happens in many quantum optical detectors , which are mostly based on the irreversible absorption of the measured radiation .
actually , even in a measurement scheme that somehow preserves the signal for further uses , one is faced by the information gain versus state disturbance trade - off , _
i.e. _ by the fact that the more information is obtained , the more the signal under investigation is being modified .
actually , the most informative measurement of an observable @xmath0 on a state @xmath1 corresponds to its ideal projective measurement , which is also referred to as von neumann _ second kind _ quantum measurement @xcite . in an ideal projective measurement
the outcome @xmath2 occurs with the intrinsic probability density @xmath3 , whereas the system after the measurement is left in the corresponding eigenstate @xmath4 .
a projective measurement is obviously repeatable , since a second measure gives the same outcome as the first one .
however , the initial state is erased , and the conditional output do not permit to obtain further information about the input signal .
the opposite case corresponds to a fully non - destructive detection scheme , where the state after the measurement can be made arbitrarily close to the input signal , and which is characterized by an almost uniform output statistics , _
i.e. _ by a data sample that provides almost no information .
besides fundamental interest , the realization of a projective measurement of the quadrature would have application in quantum communication based on continuous variables . in facts
, it provides a reliable and controlled source of optical signals . on the other hand ,
a fully non - destructive measurement scheme is an example of a quantum repeater , another relevant tool for the realization of quantum network . between these two extremes
we have the entire class of quantum nondemolition ( qnd ) measurements .
such intermediate schemes provide only a partial information about the measured observable , and correspondingly are only partially distorting the signal under investigation . in particular , in this paper
, we show how to attain an optimized qnd measurement of quadrature _ i.e. _ scheme which minimizes the information gain versus state disturbance trade - off .
most of the schemes suggested for back - action evading measurements are based on nonlinear interaction between signal and probe taking place either in @xmath5 or @xmath5 media ( both fibers and crystals ) @xcite , or on optomechanical coupling @xcite .
a beam - splitter based scheme has been earlier suggested to realize optical von neumann measurement @xcite .
here , we focus our attention on an interferometric scheme which requires only linear elements and single - mode squeezers . a schematic diagram of the suggested setup is given in fig .
[ f : setup ] .
the signal under examination @xmath6 and the probe ( meter ) state @xmath7 are given by @xmath8 where @xmath9 , @xmath10 are eigenstates of the field quadratures @xmath11 , @xmath10 of the two modes , and @xmath12 and @xmath13 are the corresponding wave - functions .
the two beams are linearly mixed in a mach - zehnder interferometer with internal phase - shift given by @xmath14 .
there are also two @xmath15 plates , each imposing a @xmath16 phase - shift .
overall , the interferometer equipped with the plates is equivalent to a beam splitter of transmittivity @xmath17 . however , the interferometric setup is preferable to a single beam splitter since it permits a fine tuning of the transmittivity . after the interferometer ,
one of the two output modes is revealed by homodyne detection , whereas the second mode is firstly displaced by an amount that depends on the outcome of the measurement ( feedback assisted amplitude modulation ) , and then squeezed according to the transmittivity of the interferometer ( see details below ) . as we will see , either by tuning the phase - shift of the interferometer , or by exciting the probe state @xmath7 in a squeezed vacuum , and by varying the degree of squeezing , the action of the setup ranges from a projective to a non - destructive measurement of the field quadrature as follows : 1 .
the statistics of the homodyne detector ranges from a distribution arbitrarily close to the intrinsic quadrature probability density of the signal state @xmath18 to an almost uniform distribution ; 2 .
the conditional output state , after registering a value @xmath19 for the quadrature of the signal mode , ranges from a state arbitrarily close to the corresponding quadrature eigenstate @xmath20 to a state that approaches the input signal @xmath6 .
the two features can be summarized by saying that the present scheme realizes the whole set of qnd measurements of a field quadrature .
in addition , the interferometer can be tuned in order to minimize the information gain versus state disturbance trade - off , _
i.e. _ to achieve an optimal qnd measurement of quadrature .
such a kind of measurement provides the maximum information about the quadrature distribution of the signal , while keeping the conditional output state as close as possible to the incoming signal .
the paper is structured as follows . in the next section
we analyze the dynamics of the measurement scheme , and describe in details the action of linear feedback and tunable squeezing on the conditional output state and on the homodyne distribution . in section
[ s : lim ] we analyze the limiting cases of strongly squeezed and anti - squeezed probes , which correspond to projective and non - destructive measurements respectively . in section [ s : opt ] we introduce two fidelity measures , in order to quantify how close are the conditional output and the homodyne distribution to the input signal and its quadrature distribution respectively . as a consequence
, we are able to individuate an optimal set of configurations that minimize the trade - off between information gain and state disturbance .
section [ s : outro ] closes the paper with some concluding remarks .
let us now describe the interaction scheme in details .
the evolution operator of the interferometer is given by @xmath21 $ ] , such that the input state @xmath22 after the interferometer the quadrature of one of the modes ( say mode @xmath23 ) is revealed by homodyne detection .
the distribution of the outcomes is given by @xmath24 \qquad \pi(x)=|x\rangle\langle x| \label{probx}\;,\end{aligned}\ ] ] @xmath25 being the povm of the homodyne detector .
since the reflectivity of the interferometer is given by @xmath26 from an outcome @xmath0 by the homodyne we infer a value @xmath27 for the quadrature of the input signal .
the corresponding probability density is given by @xmath28 \right|^2 \label{probx0}\;,\end{aligned}\ ] ] and the conditional output state for the mode @xmath29 @xmath30 the amplitude of this conditional state is then modulated by a feedback mechanism , which consists in the application of a displacement @xmath31 , with @xmath32 .
such displacing action can be obtained by mixing the mode with a strong coherent state of amplitude @xmath33 ( _ e.g. _ the laser beam also used as local oscillator for the homodyne detector , see fig .
[ f : setup ] ) in a beam splitter of transmittivity @xmath34 close to unit , with the requirement that @xmath35
an experimental implementation using a feedforward electro - optic modulator has been presented in @xcite .
the resulting state is given by @xmath36 finally , this state is subjected to a single - mode squeezing transformation @xmath37 $ ] by a degenerate parametric amplifier ( dopa ) . by tuning the squeezing parameter to a value @xmath38 and using the relation @xmath39 we arrive at the final state @xmath40\:|y\rangle \label{output}\;.\end{aligned}\ ] ] the wave - function of this conditional output state
is thus given by @xmath41}{\sqrt{\int dy\ : |\psi_{\sc s}(y)|^2 \
: \left|\psi_{\sc p}\left[\tan\phi(y - x_0)\right ] \right|^2 } } \label{outwave}\;.\end{aligned}\ ] ] eq .
( [ probx0 ] ) and eqs .
( [ output],[outwave ] ) summarize the filtering effects of the probe wave - function on the output statistics and the conditional state respectively .
for the probe mode in the vacuum state we have @xmath42 such that the homodyne distribution of eq .
( [ probx0 ] ) results @xmath43 where @xmath44 denotes convolution and @xmath45 a gaussian of mean @xmath19 and variance @xmath46 . the quadrature distribution of the corresponding output state is given by @xmath47 eqs ( [ vacprobx0 ] ) and ( [ vacond ] ) account for the noise introduced by vacuum fluctuations .
this noise can be manipulated by suitably squeezing the probe , thus realizing the whole set of qnd measurement . squeezed or anti - squeezed vacuum probes are described by the wave - functions @xmath48 where the information about squeezing stays in the requirement @xmath49 .
notice that squeezing the probe introduces additional energy in the system .
the mean photon number of the states in ( [ sqvac ] ) is given by @xmath50 .
using a squeezed vacuum probe eqs .
( [ vacprobx0 ] ) and ( [ vacond ] ) rewrites as @xmath51 eq .
( [ sqcond1 ] ) says that by squeezing the probe the statistics of the homodyne detectors can be made arbitrarily close to the intrinsic quadrature distribution @xmath52 , whereas eq .
( [ sqcond2 ] ) shows that , for any value of the outcome @xmath19 , the conditional output @xmath53 approaches the corresponding quadrature eigenstate @xmath20 . for @xmath54 the mean energy of the conditional output state @xmath53 increases , since it is approaching a quadrature eigenstate ( an exact eigenstate would have infinite energy ) . notice that this amount of energy is mostly provided by the probe state itself , rather than by the displacement and squeezing stages of the setup
the improvement in the precision due to squeezing , compared to that of a vacuum probe , can be quantified by the ratio of variances in the filtering gaussian of eqs .
( [ vacond ] ) and ( [ sqcond2 ] ) .
calling this ratio @xmath55 we have @xmath56 and thus , for squeezing not too low , @xmath57 . for an anti - squeezed vacuum probe eqs .
( [ vacprobx0 ] ) and ( [ vacond ] ) rewrites as @xmath58 eqs .
( [ sqcond3 ] ) and ( [ sqcond4 ] ) says that by anti - squeezing the probe the statistics of the homodyne detectors is approaching a flat distribution over the real axis , and correspondingly that the conditional output can be made arbitrarily close to the incoming signal , independently on the actual value of @xmath19 .
notice that , in principle , both projective and non - destructive measurements could be obtained with vacuum probe , simply by varying the internal phase - shift of the interferometer according to eqs .
( [ vacprobx0 ] ) and ( [ vacond ] ) .
however , this would affect also the _ rate _ of the events at the output ( since @xmath14 governs the transmittivity of the interferometer ) , and therefore may be not convenient from practical point of view .
on the other hand , when a fine tuning of the variances in eqs .
( [ sqcond1]-[sqcond4 ] ) is needed ( as for example in the optimization of the scheme , see next section ) it can be conveniently obtained by varying @xmath14 , without the need of varying the degree of squeezing of the probe .
so far we considered the two extreme cases of infinitely squeezed or antisqueezed probes .
now we proceed to quanti - fy the trade - off between the state disturbance and the gain of information for the whole set of intermediate cases .
there are two relevant parameters : i ) how close the output signal is to the input state , and ii ) how close the homodyne distribution is to the intrinsic quadrature probability . according to eq .
( [ output ] ) , after the outcome @xmath2 is being registered the conditional output state is given by @xmath59 .
since the outcome @xmath2 occurs with the probability @xmath60 of eq .
( [ probx0 ] ) , the density matrix describing the output ensemble after a large number of measurements is given by @xmath61 indeed , this is the state that can be subsequently manipulated , or used to gain further information on the system .
the resemblance between input and output can be quantified by the _
average state fidelity _ @xmath62 inserting eq .
( [ output ] ) in eq .
( [ avf ] ) we obtain @xmath63 where , for the squeezed vacuum probes we are taking into account , the transfer function is given by @xmath64 @xmath65 being the variance of the probe wave - function , _
i.e. _ @xmath66 for squeezed probe and @xmath67 for anti - squeezed probes .
f take values from zero to unit , and it is a decreasing function of the probe squeezing .
if also the initial signal is gaussian the fidelity results @xmath68 @xmath69 being the variance of the signal wave - function .
( [ gf ] ) interpolates between the two extreme cases of the previous section . in order to check this behavior
we evaluate ( [ gf ] ) for strong squeezing or anti - squeezing .
we have @xmath70 in order to quantify how close the quadrature probability of the input signal is to the homodyne distribution at the output we introduce the _ average distribution fidelity _ @xmath71 which also ranges from zero to one and it is an increasing function of the probe squeezing . for both gaussian signal and probe
we obtain @xmath72 and therefore @xmath73 notice that @xmath74 and @xmath75 are _ global _ figures of fidelity @xcite , _ i.e. _ compare the input and the output on the basis of the whole quantum state or probability distributions rather than by their first moments , as it happens by considering customary qnd parameters ( see for example @xcite , for a more general approach in the case of two - dimensional hilbert space see @xcite ) . as a matter of fact
the quantity @xmath76 is not constant , and this means that by varying the squeezing of the probe we obtain different trade - off between information gain and state disturbance .
an optimal choice of the probe , corresponding to maximum information and minimum disturbance , maximizes @xmath76 .
the maximum is achieved for @xmath77 , corresponding to fidelities @xmath78\simeq 86 \%$ ] and @xmath79\simeq 91\%$ ] .
notice that for a chosen signal , the optimization of the qnd measurement can be achieved by tuning the internal phase - shift of the interferometer , without the need of varying the squeezing of the probe .
for a nearly balanced interferometer we have @xmath80 : in this case the optimal choice for the probe is a state slightly anti - squeezed with respect to the signal , _
i.e. _ @xmath81 .
finally , the fidelities are equal for @xmath82 , corresponding to @xmath83=g[x_{\sc e}]\simeq 88 \%$ ] for non gaussian signals the behavior is similar though no simple analytical form can be obtained for the fidelities . in this case , in order to find the optimal qnd measurement , one should resort to numerical means @xcite .
in conclusions , we have suggested an interferometric scheme assisted by squeezing and linear feedback to realize an arbitrary qnd measurement of a field quadrature . compared to previous proposals the main features of our setup
can be summarized as follows : i ) it involves only linear coupling between signal and probe , ii ) only single mode transformations on the conditional output are needed , and iii ) the whole class of qnd measurements may be obtained with same setup either by tuning the internal phase - shift of the interferometer , or by varying the squeezing of the probe .
the present setup permits , in principle , to achieve both a projective and a fully non - destructive quantum measurement of a field quadrature . in practice , however , the physical constraints on the maximum amount of energy that can be impinged into the optical channels pose limitations to the precision of the measurements .
this agrees with the facts that both an exact repeatable measurement and a perfect state preparation can not be realized for observables with continuous spectrum @xcite .
of course , other limitations are imposed by the imperfect photodetection and by the finite resolution of detectors @xcite .
compared to a vacuum probe , the squeezed / anti - squeezed meters suggested in this paper provide a consistent noise reduction in the desired fidelity figure already for moderate input probe energy .
in addition , by varying the squeezing of the probe an optimal qnd measure can be achieved , which provides the maximum information about the quadrature distribution of the signal , while keeping the conditional output state as close as possible to the incoming signal .
99 j. von neumann , _ mathematical foundations of quantum mechanics _
( princeton univ .
press , princeton , nj , 1955 ) , pp .
442 - 445 .
m. d. levenson et al , phys .
* 57 * , 2743 ( 1986 ) .
a la porta et al , phys .
62 * , 28 ( 1989 ) .
s. pereira et al , phys .
lett . * 72 * , 214 ( 1994 ) .
j. poizat , p. grangier , phys .
lett . * 70 * , 271 ( 1993 ) .
s. r. friberg , phys .
lett . * 69 * , 3165 ( 1992 ) .
r.bruckmeier et al , appl .
b * 64 * , 203 ( 1997 )
. f. x. kartner , h. a. haus , phys .
a * 47 * , 4585 ( 1993 ) ; appl .
b * 64 * , 219 ( 1997 ) .
k. jacobs et al , phys .
a * 49 * , 1961 ( 1994 ) .
m. pinard et al , phys .
a * 51 * , 2443 ( 1995 ) .
g. m. dariano , m. f. sacchi , phys .
a 231 , 325 ( 1997 ) .
m. g. a. paris , phys .
lett . a * 217 * , 78 ( 1996 ) .
p. k. lam et al , phys .
lett . * 79 * , 1471 ( 1997 ) .
k. banaszek , phys .
lett . * 86 * , 1366 ( 2001 ) m. g. a. paris , unpublished .
d. walls and g. milburn , _ quantum optics _ springer verlag ( berlin , 1994 ) . c. fuchs , a. peres , phys .
a * 53 * , 2038 ( 1996 ) .
m. ozawa , publ .
rims kyoto * 21 * , 279 ( 1985 ) .
h. f. hofmann , phys .
a * 61 * , 033815 ( 2000 ) . | we suggest an interferometric scheme assisted by squeezing and linear feedback to realize the whole class of field - quadrature quantum nondemolition measurements , from von neumann projective measurement to fully non - destructive non - informative one . in our setup , the signal under investigation is mixed with a squeezed probe in an interferometer and , at the output , one of the two modes is revealed through homodyne detection .
the second beam is then amplitude - modulated according to the outcome of the measurement , and finally squeezed according to the transmittivity of the interferometer . using strongly squeezed or anti - squeezed probes respectively , one achieves either a projective measurement , _
i.e. _ homodyne statistics arbitrarily close to the intrinsic quadrature distribution of the signal , and conditional outputs approaching the corresponding eigenstates , or fully non - destructive one , characterized by an almost uniform homodyne statistics , and by an output state arbitrarily close to the input signal . by varying the squeezing between these two extremes , or simply by tuning the internal phase - shift of the interferometer , the whole set of intermediate cases can also be obtained . in particular , an optimal quantum nondemolition measurement of quadrature
can be achieved , which minimizes the information gain versus state disturbance trade - off . |
the isolation of a single carbon layer via micromechanical cleavage has triggered immense research activity.@xcite apart from the anomalous quantum hall effect due to chiral dirac - like quasi - particles,@xcite the finite `` universal '' dc conductivity at the neutrality point attracted major attention.@xcite for recent reviews see refs . .
the electronic properties of graphene are characterized by two nonequivalent fermi - surfaces around the @xmath0 and @xmath1-points , respectively , which shrink to two points at the neutrality point @xmath2 ( @xmath3 is chemical potential ) .
the spectrum around these two points is given by an ( almost ) isotropic energy dispersion @xmath4 with the fermi velocity @xmath5 m / s.@xcite graphene can thus be described by an effective ( 2 + 1)-dimensional relativistic field theory with the velocity of light @xmath6 replaced by the fermi velocity @xmath7.@xcite relativistic field theories in ( 2 + 1 ) dimensions were investigated long before the actual discovery of graphene@xcite and also the two values of the universal conductivities of a clean system at the neutrality point depending on whether one includes a broadening @xmath8 or not were reported then.@xcite in the first case , one obtains @xmath9,@xcite the second case yields @xmath10.@xcite interestingly , the first value is also obtained without the limit @xmath8 within the self - consistent coherent potential approximation ( cpa).@xcite we also note that the constant conductivity holds for zero temperature , only ; for finite temperature the dc conductivity is zero.@xcite if leads are attached to the graphene sample , an external broadening is introduced and the conductivity is given by @xmath11 @xcite which has been experimentally verified for samples with large aspect ratio.@xcite this is in contrast to measurements of the optical conductivity , where leads are absent and a finite energy scale given by the frequency @xmath12 of the incoming beam renders the intrinsic disorder negligible , @xmath13 .
one thus expects the universal conductivity to be given by @xmath10 , which was measured in various experiments in graphene on a sio@xmath14,@xcite sic - substrate@xcite and free hanging.@xcite also in graphene bilayer and multilayers,@xcite as well as in graphite@xcite the conductivity per plane is of the order of @xmath15 .
the above results were obtained from the kubo or laundauer formula and assumed coherent transport . also diffusive models based on the semi - classical boltzmann approach
yield a finite dc conductivity at the neutrality point .
nevertheless , the finite conductivity was found to be non - universal@xcite in contraditions to the findings of early experiments , which suggested @xmath16.@xcite we should however stress that one can still assume a certain degree of _ universality _ , since the experimental values for the conductivity are all of the order of @xmath17 . it was argued that electron - hole puddles@xcite or potential fluctuations in the substrate@xcite can account for a finite conductivity at the dirac point .
an alternative explanation of this quasi - universal behavior seen in experiments is that there is only a logarithmic dependence on the impurity concentration due to mid - gap states and therefore only in cleaner samples deviations from the universal value are seen.@xcite on the other hand , the optical conductivity is given by the universal conductivity @xmath18 for frequencies larger than twice the chemical potential @xmath3 .
it is remarkable that this universal value also holds in the optical frequency range,@xcite a result with important consequences in applications.@xcite only for frequencies @xmath19 , the sample - dependent scattering behavior of the electrons becomes important and recent experiments in show an decay of the universal conductivity with unusual large broadening around @xmath20 which can not be explained by thermal effects.@xcite moreover , the spectral weight for @xmath21 does not reach zero as would be expected due to pauli blocking , but assumes an almost constant plateau of @xmath22 for larger gate voltage .
the first calculations of the optical conductivity of graphene , using the dirac hamiltonian were done in ref .
this study was subsequently revisited a number of times , @xcite and summarized in ref .
[ ] . in these calculations
the effect of disorder was treated in a phenomenological manner , by broadening the delta functions into lorentzians characterized by constant width @xmath23 . as shown in ref .
[ ] however , the momentum states are non - uniformly broadened , with the states close to the dirac point being much more affected by the impurities than those far away from that point . in the clean limit , the exact calculation of the optical properties of graphene was considered in ref .
[ ] , a calculation recently generalized to the calculation of the optical properties of graphene antidot lattices.@xcite in this paper , we generalize the results of ref .
[ ] by considering a finite chemical potential , including the effect of charge impurities , and the scattering by phonons .
we discuss two main corrections to the clean system and calculate the optical conductivity .
first , we include the coupling of the dirac fermions to in - plane phonons , acoustical as well as optical ones .
out - of - plane phonons only have a negligible effect on the electronic properties of graphene.@xcite secondly , we include various types of disorder which give rise to mid - gap states as well as coulomb scatterers . in sec .
[ hamilt ] , we define the phonon hamiltonian , deduce the electron - phonon interaction and calculate the electronic self - energy . in sec .
[ gc ] , we discuss the green s function which is modified due to impurities and phonons .
we then present our results for dc and optical conductivity and compare it to the experiment of ref . [ ] .
we close with remarks and conclusions .
the hamiltonian , in tight binding form , for electrons in graphene is written as @xmath24 \label{tb}\end{aligned}\ ] ] where the operator @xmath25 creates an electron in the carbon atoms of sub - lattice @xmath26 , and @xmath27 does the same in sub - lattice @xmath28 .
the hopping parameter , @xmath29 , depends on the relative position of the carbon atoms both due to the presence of a vector potential @xmath30 and due to the vibration of the carbon atoms .
the vectors @xmath31 have the form @xmath32 where @xmath33 is the carbon - carbon distance . in order to obtain the current operator we write the hopping parameter as @xmath34 expanding the exponential up to second order in the vector potential @xmath30,and assuming that the electric field is oriented along the @xmath35 direction , the current operator is obtained from @xmath36 leading to @xmath37 .
the operator @xmath38 reads @xmath39\,.\end{aligned}\ ] ] the current term of the operator proportional to @xmath40 will not be used in this paper , and therefore it is pointless to give its form here.@xcite in order to describe the effect of phonons in graphene we adopt the model developed by woods and mahan@xcite and extensively used by other authors.@xcite the potential energy of the model is a sum of two terms .
the first is due to bond - stretching and reads @xmath41 ^ 2\ , , \label{v}\ ] ] where @xmath42 and @xmath43 represent the small displacements relatively to the equilibrium position of the carbon atoms in the sub - lattice @xmath26 and sub - lattice @xmath28 , respectively .
if only the term ( [ v ] ) is used a simple analytical expression for the eigen - modes is obtained.@xcite the second term of the model is due to angle deformation and has the form @xmath44 ^ 2 \right . \nonumber\\ & + & [ ( \bm\delta_1/2+\bm\delta_3)\cdot(\bm u_a(\bm r)-\bm u_b(\bm r+\bm\delta_1 ) ) + ( \bm\delta_3/2+\bm\delta_1)\cdot(\bm u_a(\bm r)-\bm u_b(\bm r+\bm\delta_3 ) ) ] ^2 \nonumber\\ & + & [ ( \bm\delta_2/2+\bm\delta_3)\cdot(\bm u_a(\bm r)-\bm u_b(\bm r+\bm\delta_2 ) ) + ( \bm\delta_3/2+\bm\delta_2)\cdot(\bm u_a(\bm r)-\bm u_b(\bm r+\bm\delta_3 ) ) ] ^2 \nonumber\\ & + & [ ( \bm\delta_1/2+\bm\delta_2)\cdot(\bm u_b(\bm r+\bm\delta_3)-\bm u_a(\bm r-\bm a_1 ) ) + ( \bm\delta_2/2+\bm\delta_1)\cdot(\bm u_b(\bm r+\bm\delta_3)-\bm u_a(\bm r-\bm a_2 ) ) ] ^2 \nonumber\\ & + & [ ( \bm\delta_1/2+\bm\delta_3)\cdot(\bm u_b(\bm r+\bm\delta_3)-\bm
u_a(\bm r-\bm a_1 ) ) + ( \bm\delta_3/2+\bm\delta_1)\cdot(\bm u_b(\bm r+\bm\delta_3)-\bm u_a(\bm r ) ) ] ^2 \nonumber\\ & + & \left .
[ ( \bm\delta_2/2+\bm\delta_3)\cdot(\bm u_b(\bm r+\bm\delta_3)-\bm u_a(\bm r-\bm a_2 ) ) + ( \bm\delta_3/2+\bm\delta_2)\cdot(\bm u_b(\bm r+\bm\delta_3)-\bm u_a(\bm r ) ) ] ^2 \right\}\,,\end{aligned}\ ] ] where @xmath45 and @xmath46 , represent the vectors defining the unit cell .
the kinetic energy has the form @xmath47 ^ 2 + \frac{m_c}2[\dot{\bm u}_b(\bm r)]^2\,,\ ] ] where @xmath48 is the carbon atom mass .
the lagrangian @xmath49 leads to an eigenproblem of the form @xmath50 , where @xmath51 and @xmath52 is the dynamical matrix reading @xmath53 with @xmath54\,,\\\end{aligned}\ ] ] where we have redefined @xmath55 as @xmath56 and @xmath57 is the momentum of the excitation .
this model can be diagonalized numerically and its spectrum is represented in fig .
[ phonon ] .
n / m and @xmath58 n / m , such that the optical frequency at the @xmath23-point is or the order of 1600 @xmath59 . also shown
are the spectra of the simplified model discussed in the text ( dashed lines ) and the effective model discussed in appendix [ aa ] ( dashed - dotted lines ) .
[ phonon ] ] although this model can easily be solved numerically , it is useful to derive from it a simple analytical model which helps in the analytical calculation of the electron - phonon problem .
to do so , we follow suzuura and ando@xcite and introduce two effective models , one for the acoustic and the other for the optical excitations , associated with the fields @xmath60 and @xmath61 , respectively .
the effective hamiltonian for the acoustic modes has the form @xmath62 with @xmath63 and @xmath64 .
the eigenmodes of this effective hamiltonian are @xmath65 and @xmath66 with polarization vectors @xmath67 and @xmath68 respectively .
the velocity of the modes is given by @xmath69 and @xmath70 the eigenmodes ( [ egma1 ] ) and ( [ egma2 ] ) are represented in fig .
[ phonon ] by dashed lines .
they can not be distinguished from the effective model described in appendix [ aa ] .
the effective hamiltonian for the optical modes is given by @xmath71 with @xmath72 , @xmath73 , and @xmath74 .
the eigenmodes are @xmath75 and @xmath76 with polarization vectors given by @xmath77 and @xmath78 respectively .
the modes ( [ egmo1 ] ) and ( [ egmo2 ] ) are represented in fig .
( [ phonon ] ) by dashed lines .
the dashed - dotted lines of the same figure are derived from an effective model , discussed in appendix [ aa ] .
we now address the question of the electron - phonon interaction .
this comes about because the hopping @xmath79 depends on the absolute distance between neighboring carbon atoms .
we therefore have @xmath80 \simeq t[a ] \nonumber\\ & + & \frac 1 a \frac { \partial t[a]}{\partial a } ( \bm u_a(\bm r)-\bm u_b(\bm r+\bm\delta))\cdot\bm\delta\ , . \label{hopping}\end{aligned}\ ] ] replacing ( [ hopping ] ) in the hamiltonian ( [ tb ] ) and introducing the fourier representation @xmath81 and @xmath82 with similar equations for @xmath83 and @xmath84 , the electron - phonon interaction has form @xmath85\cdot\bm\delta \nonumber\\ & \times&[e^{i\bm k\cdot\bm\delta}a^\dag_\sigma(\bm k+\bm q)b_\sigma(\bm k ) + e^{-i\bm k\cdot\bm\delta } b^\dag_\sigma(\bm
k)a_\sigma(\bm k-\bm q)]\,.\end{aligned}\ ] ] since we are interested in the effect of the phonons with momentum close to the @xmath23-point , the phase @xmath86 is expanded as @xmath87 . introducing the optical modes @xmath88 , the electron - phonon interaction with the optical phonon modes has the form @xmath89\ , .
\label{hop}\end{aligned}\ ] ] introducing the acoustic modes @xmath90 , the electron - phonon interaction with the acoustic phonon modes has the form @xmath91\ , .
\label{hac}\end{aligned}\ ] ] in both eqs .
( [ hop ] ) and ( [ hac ] ) the @xmath92 operators are destruction ( creation ) operators of phonons of momentum @xmath93 and polarization @xmath94 .
the electronic green s function in the dirac cone approximation has the form @xmath95[i\omega_n-\sigma(i\omega_n)]-t^2\vert \phi(\bm k)\vert^2}\ , , \label{dyson_solved}\ ] ] with @xmath96 .
the electronic self - energy shall be given by @xmath97 where @xmath98 represents the contribution due to mid - gap states ( unitary scatterers ) as well as long - range coulomb scatterers .
the self - energy @xmath99 represents the contributions due to optical and acoustic phonons .
we note that the self - energies originating from the electron - phonon and coulomb interaction are evaluated at the dirac momentum @xmath100 . in the following
, we will discuss the green s functions due to the the various contributions .
the physical origin of mid - gap states in the spectrum of graphene is varied .
cracks , edges , vacancies@xcite are all possible sources for mid - gap states . from an analytical point of view
, these types of impurities ( scatterers ) are easily modeled by considering the effect of vacancies .
we stress , however , that this route is chosen due to its analytic simplicity .
the effect of mid - gap states on the conductivity of graphene was first considered by peres _
@xcite for the case a half - filled system .
considering the effect of a local scattering potential of intensity @xmath101 , the green s function has the form of eq .
( [ dyson_solved ] ) with the retarded self - energy @xmath102 ^ 2+[\epsilon_0\pi r(\omega)]^2}\,,\ ] ] where the functions @xmath103 and @xmath104 are defined by @xmath105 mid - gap states are obtained by making the limit @xmath106 , which resembles the unitary limit . clearly , @xmath104 is the density of states per spin per unit cell .
let us write @xmath107 as a sum of real and imaginary parts , @xmath108 ( note that @xmath109 ) .
the functions @xmath103 and @xmath104 are determined self - consistently through the numerical solution of the following set of equations : @xmath110\,,\end{aligned}\ ] ] with @xmath111 and @xmath112 given by @xmath113 and @xmath114 in figure [ dos ] , we compare the density of states computed using the coherent potential approximation ( cpa ) equations with that obtained from a numerical exact method .
@xcite it is clear that the cpa captures the formation of mid - gap states in a quantitative way .
the main difference is the presence of a peak at zero energy in the exact density of states , whose measure is quantitatively negligible . and @xmath115 . here and in the following figures we use @xmath116 ev and a cutoff energy of @xmath117 ev . [ dos ]
] in figure [ cpase ] we depict the self - energy calculation using the cpa equations , for two different values of the impurity concentration .
it is clear that @xmath118 increases close to zero energy leading to a broadening of the electronic states close to the dirac point . for two impurity concentrations , @xmath119 and @xmath120 .
the left panel shows the imaginary part and the right one shows the real part of the self - energy .
[ cpase ] ] it has been argued that charged impurities are crucial to understand the transport properties of graphene on top of a silicon oxide substrate.@xcite in what follows , we compute the electronic self - energy due to charge impurities , using second order perturbation theory in the scattering potential .
electronic scattering from an impurity of charge @xmath121 leads to a term in the hamiltonian of the form @xmath122\,.\ ] ] in momentum space @xmath123 reads @xmath124,\ ] ] where @xmath125 reads @xmath126 with @xmath127 the bare and @xmath128 the full green s functions , the dyson equation due to one coulomb impurity reads @xmath129 if we consider a finite density per unit cell , @xmath130 , and incoherent scattering between impurities , the second - order self - energy is given by @xmath131 where a term of the form @xmath132 was absorbed in the chemical potential , since it corresponds to an energy shift only .
note that we have replaced @xmath125 by @xmath133 , which corresponds to include the effect of electronic screening in the calculation .
the form of @xmath133 is ( in s.i .
units)@xcite @xmath134 where @xmath135 is the silicon oxide relative permittivity , @xmath136 is the distance from the charge to the graphene plane , and @xmath137 is given by @xmath138 where @xmath139 is the self - consistent density of states as computed from the cpa calculation ( @xmath140 is the area of the unit cell ) .
the self - energy ( [ sec ] ) is dependent both on the momentum @xmath141 and on the frequency .
however , we are interested on the effect of the self - energy for momentum close to the dirac point . within this approximation
the imaginary part of the retarded self - energy becomes diagonal and momentum independent , reading ( @xmath142 ) @xmath143 the self - energy contribution due to coulomb impurities is the most relevant one in order to fit the experimental data of ref .
[ ] and is shown on the right hand side of fig .
[ s0fit ] .
following the same procedure as in the previous subsection , the self - energy due to optical phonons within first order perturbation theory is given by @xmath144 the analytical form of the self - energy due to acoustic phonons reads @xmath145 the unperturbed green s functions have the form @xmath146 ^ 2 } \,,\ ] ] and @xmath147 and @xmath148 .
the matsubara summation over the frequency @xmath149 is done using standard methods .
if we include the effect of disorder , the unperturbed green s function @xmath150 should be replaced by the dressed green s function due to the impurities .
for the ( relevant ) case of optical phonons , where the phonon dispersion is approximates by @xmath151 , the calculations are simple to do and the result for the imaginary part of the self - energy due to optical phonons is @xmath152 \,,\end{aligned}\ ] ] with @xmath153 , @xmath154 and @xmath155 and @xmath156 the bose and fermi functions , respectively .
the self - energy due to optical phonons , computed using the disordered electronic green s function , is compared with the same quantity computed using the bare electronic green s function in the central panel of fig .
it is clear that the imaginary part of the self - energy has a larger value when the disordered green s function is used .
however , in the region of frequencies @xmath157 , at @xmath158 , the imaginary part coming from the optical phonons is zero , both when one uses the bare and the disordered green s functions .
this is due to the arguments of the bose and fermi functions . .
the self - energy due to optical phonons depends on temperature and on the chemical potential @xmath3 .
we have chosen @xmath159 ev and @xmath160 k. in the central panel we show the calculation for self - energy due to optical phonons both using the bare electronic green s function ( solid line ) and the green s function with mid - gap states ( dashed line ) .
the acoustic self - energy is independent of the chemical potential and is represented for a temperature of @xmath160 k. [ se ] ] in fig . [ se ] , we depicted the self energy of the short ranged impurities together with those due to acoustic and optical phonons .
it clear that the effect of acoustic phonons is negligible at low energies .
the self - energy due to optical phonons depends on the chemical potential , and is represented for a gate voltage of @xmath161 v ( @xmath159 ev ) . to be consistent ,
also the phonon propagator has to be dressed due to its interaction with impurities .
the phonon propagator shall be renormalized within the rpa - approximation , i.e. , @xmath162 where the first order of the phononic self - energy @xmath163 is proportional to the polarization defined as @xmath164 with @xmath165 denoting the density operator .
explicitly , we get for the imaginary part of the retarded phononic self - energy @xmath166\;,\end{aligned}\ ] ] where the dimensional function @xmath167 is given in appendix [ ab ] . in fig .
[ disorder ] , the self - energy due to electron - phonon scattering ( left ) and the conductivity ( right ) are shown as they result using the bare ( dashed ) and dressed ( full ) phonon propagator .
since the effect is hardly appreciable , we show also the results where the phononic self - energy has been multiplied by a factor 10 ( dotted - dashed ) .
the renormalization of the phonon propagator due to disorder is thus negligible and the results of the following section will be obtained using the bare phonon propagator .
in this section , we discuss the transport properties of graphene due to the various sources of one - particle scattering .
this is done within the kubo formalism .
the kubo formula for the conductivity is given by @xmath168 with @xmath169 the area of the sample , and @xmath170 the area of the unit cell , from which it follows that @xmath171 and @xmath172 where @xmath173 is the charge stiffness which reads @xmath174 the function @xmath175 is obtained from the matsubara current - current correlation function , defined as @xmath176 the calculation of the conductivity amounts to the determination of the current - current correlation function .
the real part of the dc conductivity is given by @xmath177 where @xmath178 is the fermi function and @xmath179 is a dimensionless function that depends on the full self - energy . in the limit of zero temperature
the derivative of the fermi function tends to a delta - function and the conductivity is given by @xmath180 with @xmath181 given by @xmath182 and @xmath183 and @xmath173 the cutoff bandwidth . in fig .
[ s0fit ] we plot @xmath184 as function of the gate voltage , @xmath185 , considering the effects of both charged impurities , mid - gap states and acoustic phonons ( the optical phonons do not contribute to @xmath184 ) .
it is clear that the linear behavior of @xmath184 as function of @xmath185 is recovered .
the theoretical curves are plotted together with the experimental data of ref .
after having fitted the experimental curves , it is clear that the most dominant contribution to the dc conductivity is coming from coulomb impurities . ,
conductivity with the full self - energy ( solid line ) due to phonons , mid - gap states , and charged impurities , conductivity ( dashed - line ) with self - energy due to mid - gap states and charged impurities , only .
the parameters are @xmath186 k , @xmath187 , @xmath188 , and @xmath189 .
the dashed - dotted line is the same as the solid line but with @xmath190 .
right panel : imaginary part of the self - energy , @xmath191 , due to charged impurities , the most dominant contribution .
[ s0fit ] ] let us now further discuss the conductivity as function of the gate voltage @xmath185 , which relates to the chemical potential as @xmath192 . in figure [ fig_s0 ] we show @xmath193 as function of @xmath185 considering both charged impurities and short range scatterers ( left panel ) , having the same impurity concentration .
we note that the conductivity , albeit mostly controlled by charged impurities still has finger prints of the finite @xmath101 scatterers , due to the asymmetry between the hole ( negative @xmath185 ) and particle ( positive @xmath185 ) branches .
the conductivity follows closely the relation @xmath194 , except close to the dirac point where is value its controlled by the short range scattering .
if we suppress the scattering due to charged impurities , which should be the case in suspended graphene , only the scattering due to short range scatterers survive . in this case
the right panel of fig .
[ fig_s0 ] shows that there is a strong asymmetry between the hole and the particle branches of the conductivity curve even for a value of @xmath101 as large as 100 ev .
moreover , the smaller the value of @xmath101 the larger is the asymmetry in the conductivity curve .
we note that asymmetric conductivity curves were recently observed .
@xcite on the contrary , if the experimental data shows particle - hole symmetry of @xmath193 around the dirac point , then the dominant source of scattering is coming from very strong short range potentials , i.e. , scatterers that are in the unitary limit . , in units of @xmath195 , considering both short range and charged scatterers .
righ : conductivity @xmath193 when the influence of charge scatterers is removed .
[ fig_s0 ] ] the finite frequency part of the conductivity is given by @xmath196\ , , \label{sw}\ ] ] where @xmath178 is the fermi function and @xmath167 is a dimensionless function given in appendix [ ab ] .
as can be see from this appendix , the function @xmath167 depends on the self energy , which is due both to impurities and phonons .
we will first discuss the optical conductivity without coulomb scatterers since they should not be present in suspended graphene . in figure [ somega ] , we plot the conductivity of a graphene plane in units of @xmath197 .
the calculation is made at two different temperatures , @xmath160 k and @xmath198 k , and for a density of impurities @xmath199 .
the calculation compares the conductivity with and without the effect of the phonons .
the main effect induced by short - ranged impurities is the existence of a finite light - absorption in the frequency range @xmath200 .
the optical phonons increase the absorption in this frequency range .
we have checked that the effect of the acoustic phonons is negligible .
the optical phonons also induce a conductivity larger than @xmath18 for frequencies above @xmath201 .
the effect is more pronounced at low temperatures .
45 , 300 k. each panel has two curves , the solid curve is the conductivity with only short - ranged impurities , the dashed curve is the conductivity with both short - ranged impurities and phonons , with the self - energy due to phonons computed with the full green s function .
the chemical potential is that associated with a gate voltage of @xmath161 v. the vertical , dashed line marks twice this value .
[ somega ] ] in figure [ somega_ii ] , we again plot the conductivity of a graphene plane in units of @xmath197 and at temperature @xmath160 k. this time we compare different impurity densities @xmath202 ( left hand side ) and gate voltages / chemical potentials ( right hand side ) .
there is more absorption for frequencies in the region @xmath200 the larger the impurity concentration is .
this is because the number of mid - gap states is proportional to @xmath203.@xcite for larger gate - voltage a plateau is reached at frequencies in @xmath200 . for small gate voltages , @xmath185 , the absorption in the region @xmath200 is larger than for larger @xmath185 .
45 k. the left hand side shows curves for various short - ranged impurity densities @xmath202 .
the right hand side shows curves for various gate voltage @xmath185 .
the vertical , dashed lines mark twice the chemical potential .
[ somega_ii ] ] k , @xmath204 , and @xmath190 .
the dashed vertical lines correspond to @xmath205 and the shorter dotted - dashed ones to @xmath206 .
[ fig_sw ] ] .
the parameters are @xmath186 k , @xmath204 , @xmath190 , and @xmath189 .
[ swfit ] ] we will now discuss the optical conductivity with coulomb scatterers which are generally present in graphene on a substrate @xcite . in figure [ fig_sw ] , we plot the conductivity of a graphene plane in units of @xmath197 , including the effect of midgap states , charged impurities and phonons .
again , the main feature is that the conductivity is finite in the range @xmath200 and increases as the gate voltage decreases .
we choose the concentration of midgap states in fig .
[ fig_sw ] to be one order of magnitude smaller than the one of coulomb scatterers@xcite , and therefore the conductivity is mainly controlled by phonons and charged impurities . another feature of the curves in fig . [ fig_sw ] is the large broadening of the interband transition edge at @xmath205 ( indicated by vertical dashed lines ) .
note that this broadening is not due to temperature but to charged impurities , instead .
in fact , the broadening for all values of @xmath185 is larger when the conductivity is controlled by charged impurities rather than by midgap states .
the coupling to phonons produces a feature centered at @xmath207 where @xmath208 denotes the lo - phonon energy corresponding to the wave number 1600@xmath59 ( indicated by shorter dotted - dashed vertical lines ) . for gate voltages with @xmath209
there appears a similar feature at @xmath210 which is not washed out by disorder .
the optical phonons thus induce a conductivity larger than @xmath18 around these frequencies .
this effect is washed out at larger temperatures and @xmath211 .
all these effects are consistent with the recent infrared measurements of graphene on a sio@xmath14 substrate @xcite which are shown as dashed lines in fig .
[ swfit ] .
notice that there is only one fitting parameter involved which is adjusted by the conductivity curve at zero chemical potential .
whereas for low gate voltage the agreement is good , there is considerable weight missing for higher gate voltage ( @xmath212 volts ) .
nevertheless , all theoretical lines predict lower conductivity than the experimental measurements , such that our model with only one fitting parameter is consistent .
since we have included in our calculation all possible one - particle scattering mechanisms , the missing weight at large gate voltages could be attributed to electron - electron interactions in graphene , which become important at this electronic densities .
in this paper we have computed the optical conductivity of graphene at finite chemical potential , generalizing the results of ref .
the calculation includes both the effect of disorder ( mid - gap states and charged impurities , which have a different signature in the dc conductivity@xcite ) and the effect of phonons ( optical and acoustic ) .
it is shown that at low temperatures the effect of acoustic phonons in negligible , since it induces an imaginary part of the electrons self - energy that is much smaller than the imaginary part induced by the impurities . for a discussion based on the boltzmann equation ,
see ref . .
the imaginary part induced by optical phonons is of the order of the imaginary part induced by impurities .
still , optical phonons are only important in the calculation of the real part of @xmath213 , they play no role in the calculation of @xmath184 in the temperature range @xmath214 $ ] k. the self - energies due to mid - gap states and due to optical phonons are in a sense complementary , since the imaginary part coming from impurities is large at the dirac point , whereas the imaginary part coming from the optical phonons increase linearly with the energy away from the dirac point . the imaginary part of the self energy due to charged impurities is a non - monotonous function of the energy , growing first linearly but changing to a decaying behavior of the form @xmath215 , for large energies . in section [ oc ]
, we have discussed the different scattering mechanisms separately , since for suspended graphene coulomb scatterers will be absent . since vacancies ,
corresponding to an infinite potential where particle - hole symmetry is restored , are unlikely , we model the short - ranged potentials due to cracks , ripples , etc . by large , but finite
short ranged potential . for the dc conductivity , this leads to an asymmetry of the hole- and electron - doped regime .
the most notable effect on the optical conductivity of both the optical phonons and the impurities is the induction of a finite energy absorption in the energy range @xmath200 , a region where the clean theory predicts a negligible absorption .
the optical phonons also induce a conductivity larger than @xmath18 around @xmath216 .
it is interesting to note that for frequencies away from the dirac point the imaginary part of the self - energy due to optical phonons is linear in frequency , a behavior similar to that due to electron - electron interactions in graphene .
it is clear from fig .
[ swfit ] that in general the calculated absorption in the range @xmath200 is not as large as the experimental one .
it is also noticeable that for small gate voltages there is a reasonable fit of both the absorption and of the broadening of the of the step around @xmath20 .
for large values of the gate voltage the calculated absorption is smaller than the measured one .
this suggests that some additional scattering mechanism is missing in the calculation of the optical conductivity .
the missing mechanism has to be more effective at large gate voltage .
a possibility are plasmons of the type found in ref .
another possibility is that water molecules which are especially active in the infrared regime , are contributing to the missing weight . in synthesis
we have provided a complete and self - consistent description of the optical conductivity of graphene on a substrate ( including coulomb scatterers ) and suspended ( without coulomb scatterers ) .
our results are in qualitative agreement with the experimental results .
to meet a quantitative agreement further research ( also on suspended graphene ) is necessary , but water molecules underneath the graphene sheet are likely to account for the missing weight .
we thank d. n. basov , a. k. geim , f. guinea , p. kim , and z. q. li for many illuminating discussions .
we thank d. n. basov , and z. q. li for showing their data prior to publication .
we also thank vitor pereira for the exact curves we give in fig .
this work was supported by the esf science program instans 2005 - 2010 , and by fct under the grant ptdc / fis/64404/2006 .
we start by expanding the matrix elements of the dynamical matrix ( [ dm ] ) up to second order in the momentum @xmath93 .
we then introduce acoustic , @xmath217 , and optical , @xmath218 , modes .
this procedure leads to the following eigenvalue problem @xmath219 with @xmath220 with @xmath221 and @xmath222 .
further , we have @xmath223 and @xmath224 with @xmath225 and @xmath226 .
the eigenvalue problem can be put in the form @xmath227 let us first look at the acoustic modes .
they corresponds to @xmath228 . in this case
we can write @xmath229 from which an eigenvalue equation for @xmath90 follows : @xmath230 in the case of the optical modes one has @xmath231 , from which we can write @xmath232 which leads to @xmath233 the dashed - dotted lines in fig .
[ phonon ] are the eigenvalues of eqs .
( [ u ] ) and ( [ v ] ) .
in eq . ( [ sw ] ) the dimensionless @xmath167 function was introduced .
let us define @xmath234\times \nonumber\\ & & [ ( a - b)^2+(i_a+i_b)^2]\ , , \\
d_2&=&2[(a+b)^2+(i_a - i_b)^2]\times \nonumber\\ & & [ ( a+b)^2+(i_a+i_b)^2]\,.\end{aligned}\ ] ] the function @xmath167 is expressed in terms of the above auxiliary functions as follows | we compute the dc and the optical conductivity of graphene for finite values of the chemical potential by taking into account the effect of disorder , due to mid - gap states ( unitary scatterers ) and charged impurities , and the effect of both optical and acoustic phonons .
the disorder due to mid - gap states is treated in the coherent potential approximation ( cpa , a self - consistent approach based on the dyson equation ) , whereas that due to charged impurities is also treated via the dyson equation , with the self - energy computed using second order perturbation theory .
the effect of the phonons is also included via the dyson equation , with the self energy computed using first order perturbation theory . the self - energy due to phonons is computed both using the bare electronic green s function and the full electronic green s function , although we show that the effect of disorder on the phonon - propagator is negligible .
our results are in qualitative agreement with recent experiments .
quantitative agreement could be obtained if one assumes water molelcules under the graphene substrate .
we also comment on the electron - hole asymmetry observed in the dc conductivity of suspended graphene . |
mixing of scalar fields is a problem that is crucial to several environmental issues as well as engineering applications . in many situations
the underlying flow is spatially smooth and divergence - free while molecular diffusion is usually much weaker than the stirring strength of the flow ( see e.g. @xcite ) .
notwithstanding the apparent simplicity of the flow , its effect on the scalar field can be rather complex : a simple time - dependence is often sufficient for the flow to be chaotically mixing in which case the gradients of the scalar fields are greatly amplified ( @xcite ) .
@xcite recognized that this amplification is responsible for the rapid dissipation of any initial scalar inhomogeneity and thus the efficiency at which a scalar is mixed . in the continual presence of sources and sinks ,
a statistical equilibrium is attained in which the rate of injection of scalar variance balances the rate of its dissipation . in this case , the most basic way to measure the flow s mixing efficiency is to consider the equilibrium variance of the scalar : the lower its value , the better mixed is the scalar field . @xcite
derived a rigorous lower bound for the scalar variance that was further enhanced by @xcite using the scalar dissipation rate as a constraint . @xcite and @xcite derived bounds for the small- and large - scale scalar variance ( ) . this set of bounds
have successfully captured some of the key parameters in the flow and source - sink distribution that control the scalar variances .
their general applicability means that they can be used to test theoretical predictions of scalar mixing for various flow and source - sink configurations .
this is especially useful for high - pclet flows ( @xmath1 ) for which analytical solutions are difficult to obtain while high - resolution numerical simulations can become prohibitively expensive .
however , the bounds on the variance of the scalar and its gradient do not depend on the gradients of the velocity field and in many cases , can be realized by uniform flows .
they therefore do not capture the effect of stirring .
these bounds are then relevant when the mixing of a scalar is mainly controlled by the process of transport from the sources to the sinks . motivated by the apparent lack of control of the stirring process
, we here focus on the characteristic lengthscale , @xmath6 , at which the scalar variance is dissipated , or equivalently its inverse , the dissipation wavenumber , @xmath7 . its value , should , within a suitable range of parameters , be directly related to the batchelor lengthscale , @xmath8 .
the latter lengthscale , obtained in @xcite , describes the effect of stirring on the spatial structure of the scalar field .
we here examine the behaviour of @xmath0 for different values of the control parameters , @xmath2 and @xmath9 , where @xmath9 denotes the ratio of the characteristic lengthscale of the velocity , @xmath4 , and that of the source field , @xmath5 . after formulating the problem in section [ sec : formulation ] ,
we next seek a set of upper bounds for @xmath0 ( section [ sec : bounds ] ) . in section [ sec : regimes ] , we investigate the behaviour of these bounds as @xmath9 varies .
we find that , in the high - pclet limit , the behaviour of @xmath0 is characterized by four distinct regimes , one of which corresponds to the batchelor regime .
the use of homogenization theory implies a fifth regime for @xmath0 . in section [ sec : numerics ] ,
we examine the relevance of the bounds by performing a set of numerical simulations for a renewing type of flow .
we conclude in section [ sec : conclusions ] .
the temporal and spatial evolution of the concentration , @xmath10 , of a passive scalar , continually replenished by a source - sink distribution , is given by the forced advection - diffusion equation .
its general form , expressed in terms of dimensional variables , is given by [ ad - nd ] _
t(,t)+(/_u , t)(,t)= ( , t)+ s(/_s ) , where @xmath11 is the molecular diffusivity , @xmath12 is an incompressible velocity field ( i.e. @xmath13 ) and @xmath14 is a steady source field .
they are prescribed within a domain , @xmath15 , that we take to be a @xmath16-dimensional box of size @xmath17 on which we apply either periodic or no - flux boundary conditions .
this way , the boundaries can not generate any additional variability in the scalar field .
the amplitude of the velocity and source field is respectively measured by @xmath18 and @xmath19 , where @xmath20 represents a space - time average such that [ eqn : spacetime ] f _ t_0^tdt_d
f(,t ) , and @xmath21 denotes the volume of the domain . without loss of generality , we can assume that the spatial averages of @xmath22 and @xmath23 are both zero ( where negative values of @xmath24 correspond to sinks for @xmath25 ) so that @xmath10 eventually attains a statistical equilibrium with @xmath26 .
the first parameter is the pclet number , @xmath2 , defined as u _
u/ , which describes the strength of stirring relative to molecular diffusion .
the second parameter is the ratio , @xmath9 , of the velocity lengthscale , @xmath4 , to the source lengthscale , @xmath5 , defined as _
there are many ways to quantify mixing .
the simplest perhaps measure is given by the long - time spatial average of the scalar variance , which for @xmath26 , reads ^2 ^ 2 .
a scalar field is well - mixed when its distribution is nearly homogeneous i.e. has a value of @xmath27 that is small .
conversely , a badly - mixed scalar distribution is one that is inhomogeneous i.e. has a large value of @xmath27 .
the large - scale scalar variance introduced by the source at @xmath5 is transferred into small - scales where it is dissipated by molecular diffusion .
this transfer is greatly enhanced by the amplification of the scalar gradients induced by a stirring flow
. the average rate at which the scalar variance is dissipated is given by @xmath28 where : [ chi_def ] ||^2 .
we can now define the _ dissipation lengthscale _
, @xmath6 , as the average lengthscale at which the scalar variance is dissipated .
let the _ dissipation wavenumber _ , @xmath0 , denote the inverse of @xmath6 .
then , @xmath6 and @xmath0 are given by [ kd ] k_d^2 _ d^-2 = . by construction ,
the dissipation scales ( [ kd ] ) characterize the spatial variation of the scalar field and as such , provide an alternative way to quantify mixing .
the dissipation wavenumber is related ( although it is not always equal ) to the diffusive cut - off scale of the @xmath25-spectrum . for a freely decaying scalar ( i.e. @xmath29 ) ,
@xcite estimated this cut - off lengthscale to be independent of the initial configuration of the scalar field with [ batchelor ] _ _ b = , where @xmath8 stands for batchelor s lengthscale .
being independent of the source properties , @xmath8 can be used as a reference to which the value of @xmath0 can be compared for varying values of @xmath9 and @xmath2 .
multiplying equation ( [ ad - nd ] ) by @xmath25 and taking the space - time average ( [ eqn : spacetime ] ) gives the following integral constraint for @xmath30 : [ constraint1 ] = s .
thus , the average rate at which scalar variance is injected by the source at @xmath5 is equal to the average rate at which scalar variance is dissipated by molecular diffusion at small scales . using the integral constraint ( [ constraint1 ] ) , it is then straightforward to show that @xmath31 and @xmath27 are intimately related .
in particular , @xmath32 @xmath33 expresses the correlation between the scalar and source fields and takes the values between @xmath34 . for fixed values of @xmath35 and @xmath11 ,
there exist two ways to reduce the variance of the scalar : the first one relies on minimizing the correlation @xmath33 while the second one relies on maximizing the value of @xmath0 . minimizing the correlation @xmath33
can be achieved by choosing a flow that rapidly transports fluid parcels from a source region ( @xmath36 ) to a sink ( @xmath37 ) . in this configuration ,
the flow is not necessarily a stirring flow ; a uniform flow can be just as efficient in reducing @xmath33 ( see @xcite where the importance of efficient scalar transport from the sources to the sinks is highlighted for optimal mixing ) . the flow process that suppresses the scalar variance is in this case the process of _ transport_. the second way to reduce the scalar variance is to increase the value of @xmath0 .
this increase can be achieved by choosing a flow that rapidly stretches fluid parcels so that the magnitude of the scalar gradients are greatly amplified .
the flow process that suppresses the scalar variance is in this case the process of _
stirring_. thus , information about either @xmath33 or @xmath0 can provide us with some insight on the mechanisms involved in the reduction of @xmath27 . in the next session we focus on bounding the value of @xmath0 .
proper manipulation of the forced advection - diffusion equation ( [ ad - nd ] ) leads to a number of constraints that can be employed to deduce a set of upper and lower bounds for the mixing measures under consideration .
a first integral constraint is given by equation ( [ constraint1 ] ) .
following @xcite , a second integral constraint can be obtained by multiplying equation ( [ ad - nd ] ) by an arbitrary , spatially smooth ` test field ' , @xmath38 , that satisfies the same boundary conditions as @xmath39 .
space - time averaging and integrating by parts leads to [ constraint2 ] + = - s. choosing @xmath40 we first apply the cauchy - schwartz inequality on equation ( [ constraint2 ] ) to isolate @xmath27 .
we then use hlder s inequality which leads to the following lower bound for the variance @xmath27 : [ bnd_sigma ] @xmath41 where @xmath42 and @xmath43 are non - dimensional numbers that only depend on the ` shape ' of the source field and not on its amplitude or characteristic lengthscale .
explicitly they are given by [ c1 ] c_1= c_2= , where the hat symbol signifies differentiation with respect to @xmath44 .
using expressions ( [ measures ] ) and ( [ bnd_sigma](b ) ) , we obtain the following upper bound for @xmath31 : @xmath45 both bounds ( [ bnd_sigma ] ) and ( [ bnd_a ] ) were first derived in @xcite .
a new upper bound for @xmath31 can be obtained by considering the spatial and temporal evolution of the gradient of @xmath25 , [ grad ] _
t+ ( ) = -()^+ s , the average rate at which the variance of the scalar gradient is dissipated is @xmath46 where @xmath47 is defined by ||^2 .
multiplying equation ( [ grad ] ) by @xmath48 and taking the space - time average ( [ eqn : spacetime ] ) gives the following integral constraint for @xmath47 : [ constraint3 ] = - ( ) ^ + s , where the tensor @xmath49 $ ] is the symmetric part of the velocity gradient tensor[multiblock footnote omitted ] .
using hlder s inequality , the first term in ( [ constraint3 ] ) is bounded by : [ step1 ] |()^ | _ , t , | ( ) ^ | ||^2 , where @xmath50 is a unit vector so that @xmath51 . integrating by parts the second term in ( [ constraint3 ] ) and using the cauchy - schwartz inequality results in : [ step2 ] |s | |s|^2 ^ [ step ] combining the two bounds in ( [ step ] ) leads to the following upper bound for the dissipation rate of the variance of the scalar gradient : [ bnd_eta ] c_3 + c_2 , where @xmath43 and @xmath52 are non - dimensional numbers that depend on the shapes of the source and velocity field , respectively .
@xmath43 was previously defined in equation ( [ c1 ] ) and @xmath52 is defined by [ c3 ] c_3= _ , t , | ( u)^ | , where the tilde symbol signifies derivation with respect to @xmath53 . the upper bound for @xmath47 in equation ( [ bnd_eta ] ) can serve to bound @xmath0 by observing the following inequality that relates @xmath30 , @xmath27 and @xmath47 : [ useful_ineq ] = || , obtained by partial integration and application of the cauchy - schwartz inequality on the definition of @xmath30 in equation ( [ chi_def ] ) . using the definition ( [ kd ] ) of @xmath0 and the square of ( [ useful_ineq ] ) we then have [ bnd_kd_2 ] @xmath54 where the bounds ( [ bnd_eta ] ) on @xmath47 and ( [ bnd_sigma ] ) on @xmath27 were employed in order to deduce the last inequality .
the above quadratic inequality in @xmath31 yields the following upper bound for @xmath31 : [ bnd_b ] k_d^2__b^2 c_3 + , where as before , @xmath31 is normalized by the batchelor lengthscale ( [ batchelor ] )
. bound ( [ bnd_eta ] ) can further be improved for the particular case of a monochromatic source i.e. , a source that satisfies the helmholtz equation : [ harmonic ] s = -c_2 k_s^2 s. it follows that @xmath55 , where the latter is directly obtained using the integral constraint ( [ constraint1 ] ) .
substituting in equation ( [ constraint3 ] ) , bound ( [ bnd_eta ] ) becomes [ bnd_eta2 ] ( c_3 + c_2 ) . from constraint
( [ constraint1 ] ) , @xmath56 and thus equation ( [ bnd_eta2 ] ) provides a better bound for @xmath47 than equation ( [ bnd_eta ] ) . using this inequality , equation ( [ useful_ineq ] )
leads to [ bnd_b2 ] k_d^2__b^2c_3 + c_2 ^ 2 ^ -1 .
figure [ fig_1 ] shows the behaviour of the two bounds , given by eqs .
( [ bnd_a ] ) and ( [ bnd_b ] ) , for various pclet numbers , as a function of @xmath9 . for small pclet number ( @xmath57 ) ,
bound ( [ bnd_b ] ) does not improve bound ( [ bnd_a ] ) since for all values of @xmath9 it is either greater or similar to bound ( [ bnd_a ] ) . however , as the pclet number increases beyond @xmath58 values , the process of stirring becomes increasingly important and expression ( [ bnd_b ] ) can significantly improve the upper bound for @xmath59 .
this improvement depends on the value of @xmath9 .
it is only for values of @xmath60 that bound ( [ bnd_b ] ) becomes smaller than bound ( [ bnd_a ] ) and thus a better upper bound for @xmath59 .
thus , in the high - pclet limit ( @xmath61 ) , the two bounds capture different regimes of mixing that we now describe .
we first focus on @xmath60 .
the three terms inside the square root in equation ( [ bnd_b ] ) give rise to three different power - law regimes for the behaviour of the upper bound of @xmath59 . for @xmath62 , the last term inside the square root in equation ( [ bnd_b ] ) dominates .
thus , @xmath63 whence , k_d , , [ bnd_rg1 ] where sub - dominant terms have been dropped . in the case of a monochromatic source
, the validity of this regime extends to @xmath64 . for @xmath65 ,
it is the second term inside the square root in equation ( [ bnd_b ] ) that dominates . in this case ,
@xmath66 and thus the following applies for @xmath0 : k_d ( c_1c_2 ) ^ , ^ [ bnd_rg2 ] where sub - dominant terms have been dropped .
this scaling is likely to be related to the formation of boundary layers within which the scalar variance is large .
their generation is associated with regions in which the continual injection of scalar variance can not be suppressed by sweeping across the sources and sinks . @xcite
examined the case of a steady , uni - directional shear flow and a monochromatic source from where they obtained that for @xmath1 , @xmath67 .
nevertheless , we here find that regime ii is absent in the case of a monochromatic source . whether the scaling suggested by bound ( [ bnd_rg2 ] ) is realized by more complex flows and source functions than the one in @xcite or if bound ( [ bnd_b ] )
can be improved remains an open question .
the third regime appears for @xmath68 . in this case
, the first term inside the square root in equation ( [ bnd_b ] ) dominates and bound ( [ bnd_b ] ) becomes @xmath69 .
thus , in this regime , the bound for @xmath70 implies that @xmath0 and @xmath8 are inversely proportional to each other .
this relation corresponds to the prediction made in @xcite .
it follows that k_d , o(1 ) ^ , [ bnd_rg3 ] where sub - dominant terms have been dropped .
note the dependence of equation ( [ bnd_rg3 ] ) on the stirring timescale , @xmath71 .
it is therefore clear that in this regime , the dissipation wavenumber is governed by the balance between the processes of diffusion and stirring .
note that for a monochromatic source , this regime appears for @xmath72 .
when @xmath73 , the characteristic lengthscale of the source becomes larger than that of the velocity field and bound ( [ bnd_a ] ) becomes relevant . in this case ,
@xmath74 and thus k_d , o(1 ) .
[ bnd_rg4 ] thus , in this regime , both the processes of transport between the sources and sinks and diffusion control the behaviour of the dissipation wavenumber . although not captured by the two bounds , a fifth regime is expected to appear when the characteristic lengthscale of the flow is much smaller than that of the source ( @xmath75 ) . in this case , the large - scale solution to equation ( [ ad - nd ] ) is well - approximated by @xmath76 that satisfies the following equation : [ eqn : thetabar ] _
t | = | + s , where an effective diffusion operator has replaced the advective term in equation ( [ ad - nd ] ) .
the effective diffusivity tensor , @xmath77 , can be written as [ eqn : effdiff ] * k*=(+_t ) , where @xmath78 is the identity tensor and @xmath79 is a ( non - dimensional ) tensor that represents the enhancement of the diffusivity due to the flow .
it thus follows that for this range of values of @xmath9 , the dissipation wavenumber can be approximated by k_d^2= , 1 .
this approximation is obtained using @xmath80 , @xmath81 and multiplying equation ( [ eqn : thetabar ] ) by @xmath82 and space - time averaging to estimate @xmath83 .
the coefficients of @xmath79 can be rigorously obtained within the framework of _ homogenization theory _ in which multi - scale asymptotic methods are employed in order to derive the large - scale effect of the small - scale velocity field ( for derivation see review by @xcite and also @xcite in which the case of a continuously replenished scalar is examined ) . in general , the coefficients of @xmath79 depend on the value of @xmath2 with @xmath84 , where the exponent @xmath85 depends on the type of flow under consideration .
for shear flows ( taylor transport ) , @xmath86 ; for globally mixing chaotic advection flows , @xmath87 ; for cellular flows with closed field lines , @xmath88 ( see @xcite ) .
thus , depending on the value of @xmath85 , [ homo1 ] k_d~^_s^-1 , whence , [ homo ] k_d^2__b^2 ~^2 ^ -1 . for fixed value of @xmath2 ,
the above scaling increases faster in @xmath9 than the bound for @xmath70 in regime iv .
it follows that the validity of the asymptotic result ( [ homo ] ) is constrained by the upper bound ( [ bnd_a ] ) . for sufficiently high pclet values
, this is the case when @xmath89 .
based on this argument , the scalings ( [ homo1 ] ) and ( [ homo ] ) are expected to be valid at most when [ rangehomo ] \{1,^1-}. in general , the range of validity of the homogenization theory is limited to @xmath90 ( see @xcite ) . the relevance of @xmath91 was shown to be true for the mixing measures of @xcite , calculated for a particular class of steady flows ( with @xmath86 ) in @xcite and for a family of steady flows of various values for @xmath85 in @xcite . for chaotic flows
however ( @xmath87 ) the predictions of homogenization theory have been shown in @xcite to be surprisingly accurate even for @xmath92 .
we now examine how close the bounds are to the dissipation wavenumber , obtained from the solution of the forced advection - diffusion equation ( [ ad - nd ] ) . to that end
, we perform a set of numerical simulations for a passive scalar , advected by a renewing chaotic advection flow , the widely employed alternating sine flow ( e.g. @xcite ) .
this flow is explicitly given by @xmath93,\ ] ] where @xmath94 is the heaviside step function defined to be unity for @xmath95 and zero otherwise .
@xmath96 and @xmath97 are independent random angles , uniformly distributed in @xmath98 $ ] , whose value changes at each time - interval @xmath99 in order to eliminate the presence of transport barriers in the flow . this way the flow is globally mixing .
we choose a monochromatic source field that is given by [ eqn : s ] s(/_s ) = 2 s ( x/_s ) ( y/_s ) .
this source field satisfies equation ( [ harmonic ] ) and thus the two relevant bounds are ( [ bnd_a ] ) and ( [ bnd_b2 ] ) . .
we take the domain to be a doubly periodic square box whose size @xmath17 is equal to the largest of the two spatial lengthscales @xmath100 . for this flow and source fields , the coefficients @xmath42 , @xmath43 and @xmath52 , defined in eqs .
( [ c1 ] ) and ( [ c3 ] ) , are given by : [ c1c2c3 ] c_1=2 , c_2= 2 , c_3= .
in the high - pclet limit ( @xmath1 ) , the effective diffusivity tensor in equation ( [ eqn : effdiff ] ) can be calculated from the single - particle diffusivity of the velocity field ( see @xcite ) . for flow ( [ eqn : v ] ) , the enhancement diffusivity tensor , @xmath79 , is found to satisfy _
t==. employing equation ( [ homo ] ) , we can derive the following prediction for the dissipation wavenumber : [ htpred ] k_d^2__b^2 ^2 .
we solve the forced advection - diffusion equation ( [ ad - nd ] ) for flow ( [ eqn : v ] ) and source ( [ eqn : s ] ) using a pseudo - spectral method with resolution of up to @xmath101 grid points in each direction .
we consider different values of the two control parameters , @xmath9 and @xmath2 . in all simulations ,
the grid size is chosen to be smaller than the batchelor lengthscale , @xmath8 .
thus , @xmath102 . we let the simulation evolve in time until a well - observed , statistically steady state is reached .
the time - averages of all quantities of interest are thereafter calculated over several time periods @xmath99 .
figure [ fig_2 ] compares the two theoretical upper bounds , ( [ bnd_a ] ) and ( [ bnd_b2 ] ) , with the numerical values for @xmath59 . also shown is the prediction for @xmath59 , obtained from homogenization theory .
the two upper bounds combined with the prediction of homogenization theory capture the non - trivial dependence of @xmath59 on @xmath9 .
in particular , the theoretical curves and the numerical results share , for similar range of values of @xmath9 , similar slopes . however , the different scaling regimes associated with the bounds are more difficult to discern .
this is not surprising since for each power - law to clearly appear , @xmath9 needs to vary by at least an order of magnitude .
this is particularly true for the simulations corresponding to @xmath103 for which regime iii extends to a larger range of values of @xmath9 . for simulations with @xmath104 , regime iii
is limited to a smaller range of values of @xmath9 and a transition to the diffusive regime i appears , as demonstrated in the figure .
\(a ) @xmath105 and @xmath106 \(b ) @xmath107 and @xmath103 \(c ) @xmath108 and @xmath109 it is interesting to relate the variation of the dissipation wavenumber with the different spatial structures of the scalar field that result from varying @xmath9
. figures [ fig_4](a - c ) show three snapshots of @xmath10 obtained for three different values of @xmath9 , chosen for their clear representation of the different spatial structures that can be obtained .
also shown in the same figures are the time - averaged variance spectra , @xmath110 , defined in terms of @xmath111 , where @xmath112 : _ k^2 = _ t_1^t_2 _ k || < k+1 | _ ( t)|^2 dt , where @xmath113 is chosen to be sufficiently large for a steady state to have for long been established and @xmath114 .
figure [ fig_4](a ) displays the case of @xmath105 and @xmath115 .
this set of values corresponds more closely to the homogenization regime v ( see also figure [ fig_2 ] ) .
the scalar field ( left panel ) is essentially a superposition of a large - scale component that is proportional to the source field and a small - scale component that is generated by the stirring velocity field .
this superposition is clearly depicted in the spectrum of the variance ( right panel ) in which we observe that the majority of the scalar variance is concentrated on a single wavenumber which corresponds to the characteristic wavenumber of the source , @xmath116 .
small - scale fluctuations are present for @xmath117 , where @xmath118 is the characteristic wavenumber of the velocity and @xmath119 is the batchelor wavenumber beyond which the spectrum falls off exponentially .
the spectrum associated with the small - scale fluctuations exhibits a power - law scaling with an exponent that is somewhat smaller than the batchelor prediction of @xmath120 .
figure [ fig_4](b ) displays the case of @xmath107 and @xmath103 that most closely corresponds to regime iii . for this set of values ,
we observe the classical filamental structures that are obtained when stirring dominates . in this case
, the field has no memory of the functional form of the source . for @xmath121 ,
the variance spectrum is characterized by a clean power - law that behaves in agreement with the batchelor prediction .
figure [ fig_4](c ) displays the case of @xmath108 and @xmath103 which corresponds to the region of transition between regime i and iii ( recall that for source ( [ eqn : s ] ) there is no regime ii ) . in this case
, the flow is slowly varying with respect to the source . as a result ,
the scalar variance is , in the bulk of the domain , controlled by the process of sweeping between the sources and the sinks and thus its value is small .
the only exception is a small number of isolated thin , boundary layers within which the scalar field is highly varying .
these thin layers are formed in regions where the background flow is nearly stagnant so that the continual injection of scalar variance can not be suppressed by the process of sweeping . for our particular flow ( [ eqn : v ] ) the regions of zero velocity are lines that , depending on time , are either vertical or horizontal .
thus , in each half of the period ( horizontal / vertical ) thin layers of alternating sign of @xmath25 are formed .
these layers are similar to those obtained in @xcite for a steady , sine flow .
the ( horizontal / vertical ) thin layers that are formed in the first half of the period are then stretched in the second half at the same time as new ( vertical / horizontal ) thin layers are formed .
the formation of these highly - varying , thin layers yields two power - law scalings for the variance spectrum : for small wavenumbers ( @xmath122 ) , the spectrum has a positive , power - law behaviour , given by @xmath123 .
this behaviour can be deduced by noting that , for scales much larger than the source lengthscale , these structures are `` @xmath124-like '' .
conversely , for large wavenumbers - albeit smaller than the batchelor wavenumber ( @xmath125 ) , the batchelor spectrum is recovered .
in this work we obtained a set of upper bounds for the dissipation wavenumber , @xmath0 , of a continuously forced scalar field that is stirred by a spatially smooth velocity field .
we focused on the dissipation wavenumber because it provides a measure for the enhancement of mixing due to the process of stirring . unlike the freely decaying case in which stirring is the only mechanism for efficient mixing , in the forced case , transport can be as effective as stirring .
this is clear from equation ( [ measures ] ) in which it is easy to see that the scalar variance can be reduced either by increasing the dissipation wavenumber or by decreasing the correlation between the scalar and source fields .
previous investigations have considered a number of mixing measures for which a set of bounds were derived ( see @xcite ) . however , these bounds do not always distinguish between the processes of stirring and transport .
in particular , the bounds on the average variance of the scalar @xmath126 and its gradient @xmath127 do not explicitly depend on the velocity gradients .
thus , the effect of stirring is not captured . as a result , the bound for @xmath0 does not follow the scaling predicted in @xcite . with the aid of an additional constraint , we here derived a new upper bound for @xmath0 which , within a range of values of @xmath2 and @xmath9 , is , up to a constant , equal to the inverse of the batchelor lengthscale , @xmath128 . the process of stirring is thus reflected in this bound . for large pclet values , both the previous and the new bound become important , with the new bound significantly improving the previous bound for @xmath129 .
the scalings associated with these bounds suggest four different regimes for @xmath0 .
the use of homogenization theory implies a fifth regime .
the most interesting , perhaps , behaviour occurs for @xmath130 . for these range of values of @xmath9 , the scaling suggested by the upper bounds for @xmath0 transitions from a behaviour controlled by transport to a behaviour controlled by stirring .
we tested the relevance of our theoretical predictions for the particular example of the `` alternating sine flow '' and a monochromatic source .
we considered a large range of values for @xmath9 , covering more than three orders of magnitude : from @xmath75 ( in which case homogenization theory becomes relevant ) to @xmath131 ( in which case diffusion starts to dominate ) .
the theoretical results were shown in figure [ fig_2 ] to give a qualitatively good description of the non - trivial dependence of @xmath0 on @xmath9 .
we now discuss the relation of our results with a particular set of mixing measures , the so - called _ mixing efficiencies _ , @xmath132 .
these were defined in @xcite and @xcite in terms of @xmath133 , for @xmath134 , and the same variances obtained for @xmath135 satisfying ( [ ad - nd ] ) in the absence of a flow ( @xmath136 ) : [ mixeffs ] e_p .
@xmath132 are commonly larger than unity . in the high - pclet limit and for spatially
smooth source fields , they were shown to satisfy @xmath137 , for @xmath138 . using @xmath139 and equation ( [ measures ] ) into the definition for @xmath140
, we obtain that [ eqn : meas1 ] e_0 = c_4 , where @xmath141 is a non - dimensional number defined as @xmath142 ( the hat symbol denotes differentiation with respect to @xmath44 ) .
similarly , using equation ( [ kd ] ) , [ eqn : meas2 ] e_1 = c_5 , where @xmath143 is a non - dimensional number defined as @xmath144 .
thus , neither @xmath140 nor @xmath145 include separate information about @xmath0 or @xmath146 .
since we have no control over the value of @xmath33 , we can not directly compare the bounds for @xmath0 with those for @xmath140 or @xmath145 .
instead , the two sets of bounds provide complementary information . we note that from equations ( [ eqn : meas1 ] ) and
( [ eqn : meas2 ] ) , we expect that if the suppression of variance is solely due to the suppression of @xmath33 ( the case of a uniform flow ) , the two efficiencies @xmath140 and @xmath145 should scale similarly with @xmath2 . if however the suppression of the variance is due to an increase in @xmath0 , @xmath140 and @xmath145 are expected to scale differently .
the case of rough velocity fields is also very important because of its relevance to turbulent flows .
although in this case bound ( [ bnd_b ] ) can not be defined , it is still worth examining the implications of our results using simple scaling arguments at the cost of losing some of the mathematical rigor . in three - dimensional turbulent flows , the most energetic scales , @xmath147 ,
are large and control the transport between the sources and the sinks .
conversely , the smallest eddies have the largest shearing rate and control the stirring . a transition in the behaviour of @xmath0 is thus expected when bound ( [ bnd_a ] ) ( that is still valid for rough velocity fields )
intersects the batchelor scaling , @xmath148 . in terms of the reynolds number , @xmath149 ( where @xmath150 is the kinematic viscosity )
, the batchelor scale reads , [ batch2 ] _ _
b^2 ~ ^- , where we assume that @xmath151 . comparing ( [ batch2 ] ) with
bound ( [ bnd_a ] ) , we obtain that a transition occurs when @xmath152 . for @xmath153 ,
the scaling @xmath154 holds while for @xmath155 , bound ( [ bnd_a ] ) becomes smaller than @xmath128 and the scaling induced from ( [ bnd_rg4 ] ) is expected .
this prediction , however , should still be verified by numerical simulations . the authors would like to thank c.r .
doering , j.l .
thiffeault and w.r .
young for their thoughtful suggestions and two anonymous referees for their constructive comments .
a. tzella acknowledges financial support from the marie curie individual fellowship hydramitra no .
221827 as well as a post - doctoral research fellowship from the ecole normale suprieure .
the numerical results were obtained from computations carried out on the cemag computing centre at lra / ens . | we investigate the mixing properties of scalars stirred by spatially smooth , divergence - free flows and maintained by a steady source - sink distribution .
we focus on the spatial variation of the scalar field , described by the _ dissipation wavenumber _ , @xmath0 , that we define as a function of the mean variance of the scalar and its gradient .
we derive a set of upper bounds that for large pclet number ( @xmath1 ) yield four distinct regimes for the scaling behaviour of @xmath0 , one of which corresponds to the batchelor regime .
the transition between these regimes is controlled by the value of @xmath2 and the ratio @xmath3 , where @xmath4 and @xmath5 are respectively , the characteristic lengthscales of the velocity and source fields .
a fifth regime is revealed by homogenization theory .
these regimes reflect the balance between different processes : scalar injection , molecular diffusion , stirring and bulk transport from the sources to the sinks .
we verify the relevance of these bounds by numerical simulations for a two - dimensional , chaotically mixing example flow and discuss their relation to previous bounds .
finally , we note some implications for three dimensional turbulent flows . |
two dimensional spiral waves are observed in many different situations @xcite . in excitable biological tissue such as the cerebral cortex ,
the myocardium or the retina , spirals are usually precursors to serious pathological conditions @xcite .
thus there is great interest in understanding the dynamics of spiral waves , and how these are affected by model parameters and interactions with medium imperfections .
one of the most important theoretical breakthroughs in the study of spiral waves has been the adoption of techniques from group - equivariant dynamical systems @xcite .
typically the physical systems in which spirals are observed are modeled using systems of reaction - diffusion partial differential equations @xmath0 where @xmath1 is a function @xmath2 which can represent a vector of concentrations , electrical potentials , _ et cetera _ depending on the system being modeled .
the function @xmath3 models local reaction dynamics , @xmath4 is an @xmath5 matrix of diffusion coefficients , and @xmath6 are model parameters . equation ( [ rdpde1 ] ) is invariant under the full euclidean group of planar rotations and translations , @xmath7 ( in fact , they are also invariant under reflections in space , but for our purposes , this symmetry will not be useful ) . rigidly and uniformly rotating spiral waves are now well - understood to be _ rotating wave _ solutions of ( [ rdpde1 ] ) in the sense of equivariant dynamical systems @xcite ; that is , evolution in time is the same as a uniform spatial rotation of the initial condition about some point in space .
this fact has led to a model - independent understanding of many of the observed dynamical states and bifurcations of spirals , e.g. transition to meandering , resonant linear drifting @xcite .
for the purposes of applications , equation ( [ rdpde1 ] ) and the associated symmetry describes a highly idealized situation which is almost never achieved in reality : infinite planar domain which is both homogeneous and isotropic . in practice ,
media of propagation are finite in extent , so boundaries can play a significant role .
also , especially in excitable biological tissue , there are many potential sources of anisotropy and inhomogeneities .
many experiments ( both numerical and physical ) have shown that these imperfections have a significant impact on the dynamics of spiral waves @xcite using a model - independent dynamical systems formulation based on the idea of forced symmetry - breaking from @xmath7 to a proper subgroup of @xmath7 , many of these effects ( which result from inhomogeneities and/or anisotropy ) can be explained in terms of simple dynamics and bifurcations @xcite . in some cases ,
this approach has even led to some predictions which have been later confirmed experimentally @xcite .
it is this forced symmetry - breaking approach that we wish to pursue in this paper .
specifically , we will be interested in cases in which the euclidean symmetry of ( [ rdpde1 ] ) is broken by a small perturbation which preserves only the symmetry of a regular square lattice . at the cellular level , excitable biological tissues exhibit inhomogeneities in conduction via the gap junctions which separate the cells @xcite . in cases where the spatial scale of the spiral wave is comparable to the cellular scale , it is not unreasonable to assume that the conduction inhomogeneities caused by these gap junctions may have an impact on the dynamics of the spiral .
moreover , if one wants to investigate these effects , a reasonable simplifying first order approximation is to assume that the cells are distributed in a regular lattice array .
see also @xcite for further motivation . based on the results of @xcite , we expect that the lattice may have stabilising ( anchoring ) effects on the spirals and/or induce meandering
. these expectations will indeed be consequences of our results in this paper . at some phenomenological level
, it is also expected that our results may be a first step in understanding potential numerical effects induced by spatial - discretization of the domain for integrations of reaction - diffusion partial differential equations .
of course in most situations , we expect that numerical discretization effects will be negligeable if the spatial grid is fine enough
. however , coarse grids ( relative to the spatial scale of the spiral ) could have effects on the observed dynamics , and our work here could point to what these effects may be . in a specific instance
, we argue that forced lattice symmetry - breaking gives results which are consistent with numerically - observed transition from rigidly rotating wave to linearly translating waves ( with retracted tip ) as the rotation frequency of the spiral wave tends to zero @xcite .
this paper is organized as follows . in the next section
, we establish the symmetry properties we wish to consider , and derive the main center - bundle equations which represent the dynamics on perturbed relative equilibria ( i.e. rotating or travelling waves ) in the context of lattice forced symmetry - breaking . in section 3 , we recall a classic result by hale on averaging and the existence of invariant manifolds for perturbed systems , in a form suitable for our purposes .
section 4 deals with the effects of lattice symmetry - breaking perturbations on rotating spiral waves , and section 5 considers the effects of the perturbation on travelling waves .
we illustrate some of our results with numerical simulations which are presented in section 6 .
the proofs of our two main results are presented in appendices a and b.
let @xmath7 denote the set of all planar translations and rotations , parametrized as follows @xmath8 and whose action on a point @xmath9 , is given by @xmath10 where @xmath11 it thus follows that the group composition is the semi - direct product @xmath12 we will denote by @xmath13 the following subgroup of @xmath7 , which represents the symmetry of a regular square lattice @xmath14 let @xmath15 be a banach space , and suppose @xmath16 is a faithful and isometric representation of @xmath7 in the space of bounded , invertible linear operators on @xmath15 . for example ,
if @xmath15 is a space of functions with planar domain , a typical @xmath7 action is given by @xmath17 we consider an autonomous differential equation on @xmath15 of the form @xmath18 where @xmath19 is a small parameter , @xmath3 and @xmath20 satisfy the usual conditions @xcite guaranteeing that @xmath21 generates a smooth local semiflow @xmath22 on @xmath15 , and @xmath20 is bounded .
we will assume certain compatibility conditions between the evolution equation ( [ basicpde ] ) and the actions of @xmath7 and @xmath13 on @xmath15 via the representation ( [ aaction ] ) .
specifically , we assume @xmath23 and for @xmath24 , we have @xmath25 [ h1 ] hypothesis [ h1 ] states that @xmath3 in ( [ basicpde ] ) is @xmath26equivariant , but @xmath20 is only @xmath27equivariant .
we want to investigate the effects of the symmetry - breaking perturbation on a normally hyperbolic relative equilibrium to ( [ basicpde ] ) .
therefore , we will postulate that such a solution exists : _ ( existence of normally hyperbolic relative equilibrium ) _ there exists @xmath28 and @xmath29 in the lie algebra of @xmath7 such that @xmath30 holds for all @xmath31 , where @xmath32 .
we also assume that the set @xmath33 is a spectral set for the linearization @xmath34 with projection @xmath35 such that the generalized eigenspace @xmath36 is three - dimensional ( corresponding to the symmetry eigenvalues ) .
[ h2 ] in order to simplify the discussion , we will assume that the isotropy subgroup of @xmath37 in hypothesis [ h2 ] is trivial , i.e. @xmath38 assuming that all other hypothesis of the center manifold theorem of @xcite are satisfied , then for @xmath39 small enough , the dynamics of ( [ basicpde ] ) near the relative equilibrium reduce to the following ode system on the center bundle @xmath40 , ( where @xmath41 is the three - torus ) @xmath42 where @xmath43 , @xmath44 is a constant vector , @xmath45 is a constant real number , and @xmath19 is a small parameter .
the functions @xmath46 and @xmath47 are assumed to be smooth enough for our purposes , are @xmath48-periodic in @xmath49 , @xmath50 and in @xmath51 , and have the following symmetry property @xmath52 where @xmath53 as a consequence , if @xmath54 is a solution of ( [ cmeqs1 ] ) then so is @xmath55 .
the main theoretical tool underlying the analysis in the later sections is an approach which is presented in @xcite to prove the existence ( and asymptotic stability ) of invariant manifolds in systems of ordinary differential equations of the form @xmath56 where @xmath57 , @xmath58 , @xmath59 is a constant vector , @xmath60 is an @xmath61 by @xmath61 matrix of constants of block diagonal form @xmath62 where @xmath63 has all eigenvalues with negative real part and @xmath64 has all eigenvalues with positive real part . the main result we will need is theorem 2.3 ,
vii.2 of @xcite which we state here for convenience : ( ( * ? ? ?
* , theorem 2.3))consider the system ( [ ha80:eq24 ] ) , and define @xmath65 suppose that @xmath66 , @xmath67 satisfy the following conditions : 1 .
@xmath66 and @xmath67 are continuous and bounded in @xmath68 .
@xmath66 and @xmath67 are lipschitz in @xmath69 in @xmath70 , with lipschitz constants @xmath71 and @xmath72 respectively , where @xmath71 , @xmath72 are continuous and non - decreasing for @xmath73 , @xmath74 , and @xmath75 .
3 . @xmath66 and @xmath67 are lipschitz in @xmath76 with lipschitz constants @xmath77 and @xmath78 respectively , in @xmath70 , where @xmath77 , @xmath78 are continuous and non - decreasing for @xmath73 , @xmath74 , and @xmath79 .
4 . the function @xmath80 is bounded by @xmath81 for @xmath58 , @xmath82 , where @xmath81 is continuous and nondecreasing for @xmath82 , and @xmath83 . furthermore ,
since the matrix @xmath60 has all eigenvalues bounded away from the imaginary axis , there exists @xmath84 , @xmath85 such that for any real number @xmath86 , @xmath87 suppose that @xmath88 .
then the following conclusions are true for the system ( [ ha80:eq24 ] ) : there exist @xmath89 and continuous functions @xmath90 , @xmath91 , @xmath92 , which approach zero as @xmath93 , and a continuous function @xmath94\longmapsto { \bf r}^n\ ] ] with @xmath95 bounded by @xmath90 , lipschitz in @xmath69 with lipschitz constant @xmath91 , such that the set @xmath96 is an invariant set for the system ( [ ha80:eq24 ] ) .
if the functions @xmath66 and @xmath67 in ( [ ha80:eq24 ] ) are multiply periodic in @xmath69 with period vector @xmath97 , then @xmath95 is also multiply periodic in @xmath69 with period vector @xmath97 . finally , the asymptotic stability of the invariant set @xmath98 is the same as that of the trivial equilibrium point of the equation @xmath99 .
[ ha80:thm23 ] in the next sections , we will cast equations ( [ cmeqs1 ] ) in forms suitable for application of theorem [ ha80:thm23 ] , and prove the existence of hyperbolic invariant manifolds in these equations .
we will then give physical interpretations for these invariant manifolds .
when @xmath100 in ( [ cmeqs1 ] ) , then the right - hand side of the @xmath101 equation is strictly positive for all @xmath24 small enough .
consequently we may rescale time along orbits of ( [ cmeqs1 ] ) so that the resulting equations are of the form @xmath102 where @xmath103 is a constant vector and @xmath104 is smooth and satisfies the symmetry properties ( [ symprop ] ) .
a simple change of variables transforms ( [ pertrot ] ) into @xmath105 where @xmath106 satisfies the symmetry properties ( [ symprop ] ) .
when @xmath107 in ( [ pertrot2 ] ) , all solution curves @xmath108 are @xmath48-periodic in time on the three - torus . in physical space , these motions represent a spiral wave which is rigidly and uniformly rotating around a fixed point in the plane . the initial conditions @xmath109 and @xmath110 for any solution of ( [ pertrot2 ] ) can be thought of as parametrizing all possible centers of rotation in the plane for the spiral wave ( modulo the unit square ) , and initial phase of rotation of the spiral wave , respectively .
our main result in this section will be to show that , under certain hypotheses on @xmath111 , when @xmath112 , ( [ pertrot2 ] ) can admit hyperbolic periodic solutions , and hyperbolic invariant two - tori , where one of the frequencies of the two - tori is @xmath113 .
we note using standard results from the theory of averaging that there exists a periodic ( in @xmath51 ) near - identity change of variables which transforms ( [ pertrot2 ] ) into @xmath114 where @xmath115 is smooth , multiply periodic , satisfies ( [ symprop ] ) , and @xmath116 a simple computation shows that @xmath117 the @xmath118-equivariant planar ordinary differential equation @xmath119 has an equilibrium at @xmath120 , and any non - trivial equilibria occur in conjugate sets @xmath121 .
also , if @xmath122 is a @xmath123-periodic orbit for ( [ z4eq ] ) , then either @xmath124 is a distinct periodic orbit for @xmath125 , or @xmath126 coincides with the orbit @xmath127 , in which case the periodic solution has one of the spatial - temporal symmetries @xmath128 we are now ready to state the main result of this section , whose proof is given in appendix a. consider the @xmath118-equivariant ode ( [ z4eq ] ) . 1 .
suppose ( [ z4eq ] ) has an equilibrium point @xmath129 which is linearly stable ( resp .
unstable ) .
then for all @xmath112 small enough , ( [ avpertrot ] ) has linearly stable ( resp .
unstable ) @xmath48-periodic orbits of the form @xmath130 where @xmath131 approaches @xmath132 as @xmath93 . furthermore ,
if @xmath133 , then the periodic orbit is such that @xmath134 2 .
suppose @xmath135 is a non - trivial @xmath123-periodic orbit for ( [ z4eq ] ) , and that this orbit is linearly stable ( resp .
unstable ) .
then for all @xmath24 small enough , ( [ avpertrot ] ) has linearly stable ( resp .
unstable ) invariant two - tori of the form @xmath136\times [ 0,2\pi ] , \end{array}\ ] ] where @xmath137 approaches @xmath132 as @xmath93 .
furthermore , if @xmath129 satisfies the spatio - temporal symmetry property ( [ spatiotemporalsym ] ) then the invariant torus is such that @xmath138 [ thmrotwave ] * interpretation * as mentioned at the beginning of this section , when @xmath107 the dynamics of ( [ pertrot2 ] ) are particularly simple : all solutions are @xmath48-periodic in time on the three - torus . in physical space
, these motions represent a spiral wave which is rigidly and uniformly rotating around a fixed point in the plane .
the initial conditions @xmath109 and @xmath110 for any solution of ( [ pertrot2 ] ) can be thought of as parametrizing all possible centers of rotation in the plane for the spiral wave ( modulo the unit square ) , and initial phase of rotation of the spiral wave , respectively .
thus in the fully euclidean case , there are no preferred centers of rotation .
all points in physical space can be centers of rotation provided the proper initial conditions are chosen .
theorem [ thmrotwave ] characterizes the existence of ( generically isolated ) periodic solutions and invariant two - tori for ( [ pertrot2 ] ) and the local structure of the dynamics around these isolated periodic solutions and tori for @xmath24 .
in physical space , these hyperbolic periodic solutions and two - tori correspond to `` anchored '' rotating waves or meandering waves ( in the asymptotically stable case ) , or `` repelling '' rotating waves or meandering waves ( in the unstable case ) @xcite . therefore , whereas rotating spiral waves in the fully euclidean case have no preferred center of rotation , in the perturbed case , we expect generically that certain points in physical space ( called _ ar - points _ herein ) will act as local attractors or repellants for the spiral core . to any such ar - point in physical space
, there correspond in fact infinitely many conjugate ar - points related to each other by the lattice symmetry ( [ latticesymmetry ] ) .
perhaps counter - intuitively , a given ar - point need not coincide with the physical location of localized heterogeneities ( i.e. lattice points ) , as was shown in @xcite .
although if it does , then the solution possesses residual spatio - temporal symmetry .
the phenomenon is illustrated with numerical simulations in section 5 below .
to summarize , theorem [ thmrotwave ] states that the effects of lattice symmetry - breaking on a rotating spiral wave in a euclidean system will be to create hyperbolic rotating waves and/or hyperbolic meandering waves . in the latter case ,
one of the meander frequencies is very small ( of the order of the `` size '' of the perturbation ) . in certain cases ( where the ar - point of the wave coincides with a point of the lattice )
, the rotating wave or meandering wave has spatial temporal symmetry : for the rotating wave , rotation in physical space by @xmath139 around the lattice point is the same as time - evolution by a quarter period . in the meandering case
, the full meander pattern will have symmetry by rotation of @xmath139 around the lattice point .
all of these possibilities are illustrated in the numerical examples in section 6 .
when @xmath107 in ( [ cmeqs1 ] ) , these equations represent travelling waves when @xmath140 . in physical space , these travelling waves manifest themselves as a spiral wave that has completely unwound , with retracting tip @xcite . on the three torus ,
the corresponding solution curves are of the form @xmath141 where @xmath109 and @xmath110 are constants .
one can view these solutions as inducing a partition of the three - torus into a one - parameter family ( parametrized by @xmath110 ) of invariant two - tori , and on each of these invariant two - tori , the flow is either dense or periodic , depending on whether or not the components of the vector @xmath142 are linearly independent over the rationals .
we now wish to give a characterization of what happens to this picture when @xmath143 is small .
we have the following , which is proved in appendix b : for the function @xmath47 and the constant vector @xmath144 in ( [ cmeqs1 ] ) , define @xmath145 suppose @xmath146 is such that 1 .
@xmath147 , 2 .
@xmath148 .
further suppose that the vector @xmath149 is such that the following diophantine condition is satisfied for some constants @xmath150 and @xmath85 : @xmath151 then for all @xmath143 sufficiently small , the system ( [ cmeqs1 ] ) has a hyperbolic invariant two - torus @xmath152 such that @xmath153 as @xmath93 .
this two - torus is locally asymptotically stable ( resp .
unstable ) if @xmath154 ( resp .
@xmath155 ) . [ thmtravwave ] * interpretation *
the existence of stable travelling waves ( when the conditions of theorem [ thmtravwave ] are met ) is of some interest .
suppose that @xmath156 is such that @xmath147 and @xmath157 .
then there exists @xmath158 such that for all @xmath159 , system ( [ cmeqs1 ] ) has a hyperbolic invariant two - torus @xmath152 which represents a travelling wave . by normal hyperbolicity
, we expect that for all @xmath159 , there exists an interval @xmath160 such that ( [ cmeqs1 ] ) has a hyperbolic invariant two - torus @xmath161 for all @xmath162 .
thus , in the @xmath39 @xmath163 parameter space of ( [ cmeqs1 ] ) , there exists a region @xmath164 which resembles qualitatively figure [ retreg ] ( see also figure 4 of @xcite for an analogous phenomenon ) with the following properties : * outside @xmath164 where @xmath163 dominates @xmath39 , the results of section 4 apply and there are no stable travelling waves for ( [ cmeqs1 ] ) * on the boundary of @xmath164 , stable travelling waves ( invariant two - tori ) are created via saddle - node bifurcation , * in the interior of @xmath164 where @xmath39 dominates @xmath163 , theorem [ thmtravwave ] applies and there may exist transverse zeros of the function @xmath165 which correspond to invariant two - tori @xmath166 for ( [ cmeqs1 ] ) .
each such stable two - torus corresponds to an asymptotically stable travelling wave solution of ( [ cmeqs1 ] ) .
in which stable travelling waves of system ( [ cmeqs1 ] ) exist.,width=307 ] consider now a physical experiment on spiral waves where euclidean symmetry is broken by a weak perturbation which preserves lattice symmetry ( i.e @xmath39 is fixed , small and positive ) .
suppose an experimental parameter @xmath167 is varied continuously in such a way that in the @xmath101 equation of ( [ cmeqs1 ] ) , @xmath168 decreases as illustrated in the dashed line of figure [ retreg ] .
the following may be observed in physical space . as @xmath163 decreases
while remaining above the top boundary of @xmath164 , the spiral tip will undergo a periodic motion .
however , when @xmath163 crosses the boundary of @xmath164 , the wave will stop rotating and start drifting linearly , due to a saddle - node bifurcation of zeros in the function @xmath165 .
moreover , this linear motion will persist as the experimental parameter @xmath167 is further varied while @xmath163 stays in the interior of @xmath164 .
see also @xcite for a similar scenario in the case of rotational symmetry - breaking for spiral waves .
the above scenario is reminiscent of the experimentally observed phenomenon of retracting spiral waves . in certain numerical experiments on spiral waves @xcite
, it has been observed that as an experimental parameter is varied , the rotation frequency of a rotating spiral wave decreases , the spiral core unwinds and the radius of the circular motion grows unbounded , eventually reaching a state where the spiral is completely unwound , and the tip drifts linearly ( retracts ) when the rotation frequency approaches zero .
as the experimental parameter is further varied , the spiral remains unwound and the linearly drifting state persists in an asymptotically stable manner . in @xcite
, it was shown that the euclidean equivariant equations ( [ cmeqs1 ] ) with @xmath107 are in partial agreement with the retracting wave scenario . indeed , suppose that the dashed line in figure [ retreg ] were drawn a bit more to the left , coinciding with the @xmath163-axis ( i.e. @xmath107 for the euclidean case ) .
then , as is shown in @xcite , as @xmath163 decreases , the circular spiral tip motion grows unbounded like @xmath169 .
however , since the intersection of the region @xmath164 with the @xmath163-axis consists solely of the point @xmath140 , the analysis predicts that the linearly translating wave exists only for @xmath140 ( i.e. it is not asymptotically stable , and does not persist under further variation of the parameter ) .
this prediction is inconsistent with the above - described experiments on retracting spirals @xcite .
if one considers a situation where euclidean symmetry is broken ( i.e. @xmath24 ) , then , as our analysis reveals , asymptotically stable travelling waves near @xmath140 are possible .
we note that the pde model which is integrated in @xcite is fully euclidean symmetric ; however , the spatial discretization schemes which are used to perform the numerical integrations break the symmetry but preserve a lattice symmetry . we qualify these remarks with two important caveats .
first , if indeed euclidean symmetry - breaking ( e.g. of a numerical type ) were responsible for the experimentally observed retracting tip phenomenon ( we are not claiming that it is ) , then the rotation of the spiral just before the transition to linear translation would be markedly non - uniform . as far as we are aware , this non - uniformity is not observed .
second , as was the case in @xcite , the center bundle equations ( [ cmeqs1 ] ) predict that as @xmath163 is further varied ( beyond the region @xmath164 ) , the spiral should start rotating in the opposite direction .
this does not appear to be what happens in the experiments : as the experimental parameter is further varied , the excitability of the medium is eventually diminished to an extent where wave propagation is no longer possible @xcite .
we have performed numerical simulations to illustrate the theoretical results presented in the previous sections .
we have used a finite - difference scheme ( @xmath170 spatial grid ) with explicit time - stepping to integrate the fitzhugh - nagumo reaction diffusion pde for the inhomogeneous terms @xmath171 and @xmath172 in ( [ rdpdesim]).,width=508 ] ( left ) and @xmath173 ( right ) at some instant in time for the integration of ( [ rdpdesim ] ) in the inhomogeneous case .
one can visually observe the background effects of the lattice perturbation on these fields.,title="fig:",width=307 ] ( left ) and @xmath173 ( right ) at some instant in time for the integration of ( [ rdpdesim ] ) in the inhomogeneous case .
one can visually observe the background effects of the lattice perturbation on these fields.,title="fig:",width=307 ] ) with coefficients ( [ firstexp ] ) .
the black crosses and the red circles denote the positions of the minima and maxima ( respectively ) of the first spatial harmonic of the inhomogeneous terms in ( [ rdpdesim]).,width=624 ] ) as in previous figure , with transient removed .
we can see that the final ( steady ) state is anchored around a local maximum of the inhomogeneity field ( lattice point).,width=624 ] ) with coefficients ( [ secondexp ] ) .
the initial position of the spiral tip for each integration is labeled as i.c.1 through i.c.5 .
initial conditions i.c.1 through i.c.4 are conjugate by rotations of 90 degrees , whereas i.c.5 is not related by symmetry to the other initial conditions .
the final ( stable ) states are shown in yellow .
the system displays multistability between families of rotating waves which are anchored around non - lattice points , and spatio - symmetric meandering waves , anchored at ( dual ) lattice points.,width=432 ] ) with coefficients ( [ thirdexp ] ) .
the initial position of the spiral tip for each integration is labeled as i.c.1 through i.c.4 . which are conjugate by rotations of 90 degrees .
the final ( stable ) states are shown in yellow , and correspond to meandering waves which are anchored at non - lattice points.,width=432 ] ) with data ( [ simdata ] ) represented on a portion of the three - torus ( opposite faces in @xmath174 are identified ) .
after a transient approach , the solution curve fills out densely an invariant surface near the two - torus @xmath175 , as predicted by theorem [ thmtravwave].,width=624 ] @xmath176 { \displaystyle\frac{\partial v}{\partial t}}&=&{\displaystyle\frac{3}{10}\left(u+0.6 - 0.5v\right)+\varepsilon\,g_2(x , y ) } \end{array } \label{rdpdesim}\ ] ] on the square grid @xmath177 ^ 2 $ ] with neumann boundary conditions . with the chosen parameters , in the fully euclidean case @xmath107 , system ( [ rdpdesim ] ) admits rigidly rotating spiral waves .
we therefore integrate ( [ rdpdesim ] ) with @xmath107 long enough so the system eventually achieves a fully - formed , uniformly rigidly rotating spiral wave .
we then use this state as an initial condition for the inhomogeneous case @xmath178 in further integrations of ( [ rdpdesim ] ) .
it should be noted that for all integrations performed in this section , the `` spiral tip '' is defined as the intersection of the @xmath179 and @xmath180 contours ( which is computed numerically ) , but the results would remain qualitatively the same for any other commonly used definition of spiral tip ( e.g. maximum of @xmath181 ) @xcite .
the inhomogeneous terms @xmath171 and @xmath172 in ( [ rdpdesim ] ) are chosen to have the lattice symmetry generated by translations of @xmath182 along @xmath76 and @xmath183 axes , and rotation by 90 degrees around the origin .
specifically , both @xmath184 and @xmath185 are expressions of the form @xmath186 with the constant coefficients @xmath187 , @xmath188 and @xmath189 being changed from one experiment to the next .
the first harmonic @xmath190 is plotted in figure [ inhfield ] .
this setup could represent phenomenologically a spatially periodic ( along a lattice ) array of reaction inhomogeneities in the medium of propagation of the spiral
. equations ( [ rdpdesim ] ) are integrated until a steady ( anchored ) state is observed ( either periodic or meandering ) . in figure
[ uvfields ] , we show plots of both the @xmath191 field and the @xmath192 field at some intermediate instant in time along the integration
. figures [ anctrans ] and [ ancsteady ] show the motion in physical space of the spiral tip path ( both transients , and steady regimes ) for a first choice of coefficients @xmath187 , @xmath188 and @xmath189 in ( [ lincombcoeffs ] ) : @xmath193 where we see that the spiral gets anchored to the site of a local maximum for the first spatial harmonic of the inhomogeneity field ( i.e. anchored to a lattice point ) . in figure
[ anc4 ] , we show the motion in physical space of the spiral tip path for the following choice of coefficients @xmath187 , @xmath188 and @xmath189 in ( [ lincombcoeffs ] ) @xmath194 and for five different integrations using conjugate initial conditions .
four of the initial conditions are conjugate by rotations of 90 degrees , whereas the fifth initial condition is non - conjugate .
as predicted in section 4 , we see that the system displays multistability between conjugate families of stable ( anchored ) rotating waves , where the point of anchoring is not a lattice point , and meandering waves which are anchored at ( dual ) lattice points .
finally , in figure [ ff ] , we show the motion in physical space of the spiral tip path for the coefficients @xmath195 in ( [ lincombcoeffs ] ) , with four conjugate initial conditions . in this case , the stable steady - states are conjugate families of meandering waves anchored at non - lattice points . in order to illustrate theorem [ thmtravwave ] , we have performed a numerical simulation of the following system of the form ( [ cmeqs1 ] ) with : @xmath196 f^{\varphi}(\psi,\varphi,\varepsilon)=2\sin(4\varphi)+\cos(7\psi_1 + 6\psi_2)+\cos(6\psi_1 - 7\psi_2)\\[0.15 in ] { \displaystyle f^{\psi}(\psi,\varphi,\varepsilon)=\left(\begin{array}{c } \sin(5\varphi)\sin(\psi_1+\psi_2)+\cos(5\varphi)\sin(\psi_1-\psi_2)\\ \cos(2\varphi)(\cos(2\psi_1 + 3\psi_2)-\cos(3\psi_1 - 2\psi_2 ) ) \end{array}\right)}. \end{array } \label{simdata}\ ] ] with this data , the function @xmath165 in ( [ mdef ] ) becomes @xmath197 and we note in particular that @xmath198 is such that @xmath147 and that @xmath199 .
furthermore , the components of the vector @xmath200 are rationally incommensurable , and since the perturbations terms @xmath47 and @xmath201 are trigonometric polynomials , the diophantine condition ( [ diophantine ] ) is satisfied .
theorem [ thmtravwave ] then predicts the existence of a locally asymptotically stable invariant surface ( two - torus ) @xmath202 close to @xmath203 .
figure [ invs ] illustrates the result of a numerical simulation of the above system , and we can clearly see the transient approach of the solution to the predicted invariant surface . it should be noted that this figure represents one integration of the above system for one particular choice of initial condition , and that the trajectory fills out the invariant two - torus densely .
* acknowledgments * this research is partly supported by the natural sciences and engineering research council of canada in the form of a discovery grant ( vgl ) , and an undergraduate student research award ( lc ) .
funding from the ontario graduate scholarship ( ogs ) program is also acknowledged ( lc ) .
* appendix *
we first prove item ( i ) related to equilibria of ( [ z4eq ] ) .
suppose @xmath129 is an equilibrium point for ( [ z4eq ] ) , and write @xmath204 then upon dropping the hats , ( [ avpertrot ] ) becomes @xmath205 where @xmath206 is smooth and periodic in @xmath51 . by applying theorem [ ha80:thm23 ] in the case @xmath207 and @xmath208 , we get that there exists a periodic solution of ( [ avpertrotcentered ] ) which is a graph @xmath209 where @xmath210 as @xmath39 goes to zero , and using ( [ variation1 ] ) we get a periodic solution of ( [ avpertrot ] ) @xmath211 the stability of this periodic solution is the same as that of the equilibrium point @xmath129 in the planar system ( [ z4eq ] ) . for the conjugate equilibrium @xmath212 of ( [ z4eq ] ) , we replace ( [ variation1 ] ) with @xmath213 and set @xmath214 in ( [ avpertrot ] ) . because of the equivariance properties ( [ symprop ] ) , upon dropping the hats , we get @xmath215 which is essentially identical to ( [ avpertrotcentered ] ) .
so we have a periodic solution @xmath216 for ( [ avpertrot ] ) . if @xmath217 , then ( [ persol1 ] ) and ( [ persol2 ] ) are distinct periodic solutions of ( [ avpertrot ] ) .
however , if @xmath133 , then ( [ persol1 ] ) and ( [ persol2 ] ) are the same solution , and we thus get @xmath218 which establishes ( [ stsp ] ) .
now , we will prove item ( ii ) related to periodic orbits of ( [ z4eq ] ) .
let @xmath219 be a @xmath123-periodic solution of ( [ z4eq ] ) .
then it is well - known @xcite that near this periodic orbit , there exist local coordinates @xmath220 defined by @xmath221 which transforms ( [ avpertrot ] ) into @xmath222 where all functions are smooth , @xmath123-periodic in @xmath223 and @xmath48-periodic in @xmath51 , and where the function @xmath224 is such that @xmath225 define @xmath226 then @xmath227 ( resp .
@xmath228 ) if the limit cycle @xmath219 is linearly stable ( resp .
unstable ) .
the periodic change of variables @xmath229 transforms ( [ pertavg1 ] ) into @xmath230 where all functions are smooth , @xmath123-periodic in @xmath223 and @xmath48-periodic in @xmath51 .
system ( [ pertavg2 ] ) is in the proper form for application of theorem [ ha80:thm23 ] in the case @xmath231 and @xmath232 .
we conclude that there is an invariant two - torus for ( [ pertavg2 ] ) which is the graph @xmath233 where @xmath234 as @xmath93 . from ( [ cov1 ] ) and ( [ cov2 ] )
, we conclude that ( [ avpertrot ] ) has an invariant two - torus @xmath235 where @xmath236 we use an argument similar to that used above ( for equilibria ) to get the conjugate invariant two - tori , and the symmetry property ( [ stsp2 ] )
. it will be convenient to rewrite ( [ cmeqs1 ] ) with @xmath140 as @xmath237 where @xmath238 .
then for @xmath143 , the change of variables @xmath239 transforms ( [ cmeqs3 ] ) into the following ( upon dropping the hats ) : @xmath240 a taylor expansion of these equations yields @xmath241+\varepsilon^{\frac{3}{2 } } q_1(\psi,\varphi,\varepsilon)\\&&\\ \dot{\varphi}&=&\sqrt{\varepsilon}f^{\varphi}(\psi,\varphi^*,0)+\varepsilon f^{\varphi}_{\varphi}(\psi,\varphi^*,0)\varphi+\varepsilon^{\frac{3}{2}}[f^{\varphi}_{\varepsilon}(\psi,\varphi^*,0)+\frac{1}{2}f^{\varphi}_{\varphi\varphi}(\psi,\varphi^*,0)\varphi^2]+\\[0.1 in ] & & \varepsilon^2 s_1(\psi,\varphi,\varepsilon ) \end{array } \label{cmeqs2}\ ] ] the diophantine condition ( [ diophantine ] ) ensures that for any smooth function @xmath242 , if we denote by @xmath243 then there exists a smooth solution @xmath244 to the equation @xmath245 with @xmath246 . recall that @xmath247 . consequently
, if @xmath248 is such that @xmath249 then the near - identity change of variables @xmath250 transforms ( [ cmeqs2 ] ) into the following ( upon dropping the hats ) @xmath251+\varepsilon^{\frac{3}{2 } } q_2(\psi,\varphi,\varepsilon)\\&&\\ \dot{\varphi}&=&\varepsilon [ f^{\varphi}_{\varphi}(\psi,\varphi^*,0)-d_{\psi}{\cal y}(\psi)r'_{\varphi^*}v]\varphi+\varepsilon^{\frac{3}{2}}[f^{\varphi}_{\varphi}(\psi,\varphi^*,0){\cal y}(\psi)+f^{\varphi}_{\varepsilon}(\psi,\varphi^*,0)+\\[0.1 in ] & & \frac{1}{2}f^{\varphi}_{\varphi\varphi}(\psi,\varphi^*,0)\varphi^2-d_{\psi}{\cal y}(\psi)(r'_{\varphi^*}v{\cal y}(\psi)+g^{\psi}(\psi,\varphi^*,0)+\frac{1}{2}r''_{\varphi^*}v\varphi^2)]+\varepsilon^2 s_2(\psi,\varphi,\varepsilon ) . \end{array } \label{cmeqs4}\ ] ] if we rescale @xmath252 , ( [ cmeqs4 ] ) simplifies as @xmath253+\varepsilon^{\frac{3}{2 } } q_3(\psi,\varphi,\varepsilon)\\&&\\ \dot{\varphi}&=&\varepsilon [ f^{\varphi}_{\varphi}(\psi,\varphi^*,0)-d_{\psi}{\cal y}(\psi)r'_{\varphi^*}v]\varphi+\varepsilon [ f^{\varphi}_{\varphi}(\psi,\varphi^*,0){\cal y}(\psi)+f^{\varphi}_{\varepsilon}(\psi,\varphi^*,0)+\\[0.1 in ] & & -d_{\psi}{\cal y}(\psi)(r'_{\varphi^*}v{\cal y}(\psi)+g^{\psi}(\psi,\varphi^*,0))]+\varepsilon^{\frac{3}{2 } } s_3(\psi,\varphi,\varepsilon ) . \end{array } \label{cmeqs5}\ ] ] a final near - identity change of variables @xmath254,\,\,\,\,\,\,\,\,\,\,\,\,\,\ , \psi=\hat{\psi}+\varepsilon { \cal z}_3(\hat{\psi})\ ] ] transforms ( [ cmeqs5 ] ) into the following ( upon dropping the hats ) @xmath255+\varepsilon^{\frac{3}{2 } } q_4(\psi,\varphi,\varepsilon)\\&&\\ \dot{\varphi}&=&\varepsilon [ \mu\varphi+\lambda]+\varepsilon^{\frac{3}{2 } } s_4(\psi,\varphi,\varepsilon ) \end{array } \label{cmeqs6}\ ] ] where we recall that @xmath148 , and we define @xmath256 and @xmath257 translation of the variable @xmath51 gives the final system @xmath258+\varepsilon^{\frac{3}{2 } } q(\psi,\varphi,\varepsilon)\\&&\\ \dot{\varphi}&=&\varepsilon \mu \varphi + \varepsilon^{\frac{3}{2 } } s(\psi,\varphi,\varepsilon ) \end{array } \label{cmeqs7}\ ] ] system ( [ cmeqs7 ] ) satisfies all the hypotheses of theorem [ ha80:thm23 ] above ( i.e. theorem 2.3 , vii.2 of @xcite ) , and so we conclude that ( [ cmeqs7 ] ) admits an invariant two - torus @xmath259 ( asymptotically stable ( resp
. unstable ) if @xmath154 ( resp .
@xmath260 ) ) which corresponds to a hyperbolic invariant two - torus @xmath261 ( with @xmath153 as @xmath93 ) for ( [ cmeqs1 ] ) .
this concludes the proof of the theorem . | spiral waves in two - dimensional excitable media have been observed experimentally and studied extensively .
it is now well - known that the symmetry properties of the medium of propagation drives many of the dynamics and bifurcations which are experimentally observed for these waves . also , symmetry - breaking induced by boundaries , inhomogeneities and anisotropy have all been shown to lead to different dynamical regimes as to that which is predicted for mathematical models which assume infinite homogeneous and isotropic planar geometry .
recent mathematical analyses incorporating the concept of forced symmetry - breaking from the euclidean group of all planar translations and rotations have given model - independent descriptions of the effects of media imperfections on spiral wave dynamics .
in this paper , we continue this program by considering rotating waves in dynamical systems which are small perturbations of a euclidean - equivariant dynamical system , but for which the perturbation preserves only the symmetry of a regular square lattice . |
at the 2012 varenna summer school on _ physics of complex colloids _ , i gave a series of lectures on computer simulations in the context of complex liquids .
the lectures were introductory , although occasionally , i would mix in a more general cautionary remark
. it seemed to me that there was little point in writing a chapter in the proceedings on ` introduction to computer simulations ' .
books on the topic exist .
however , i did not quite know what to write instead .
then , over lunch , _ wilson poon _ suggested to me to write something on the limitations of existing simulations methods : where do they go wrong and why ?
i liked the idea very much .
the scope of the present manuscript is a bit broader : after a fairly general ( but brief ) introduction , i will discuss three types of issues : 1 .
computer simulation methods that seem simple yet require great care 2 .
computer simulation methods that seem reasonable but are not 3 . myths and misconceptions not all issues that i list are of direct relevance for soft matter . however , i hope that the reader will forgive me .
i should also point out that many of the issues that i discuss are very well known
sometimes they are even trivial .
however , i thought it better to list even the trivial examples , rather than assume that every single one of them is well known to all readers .
some of the issues that i highlight may not be well known , simply because i am mistaken or i have missed a key reference .
if so , i apologise .
i also apologise for the rather black - or - white way in which i present problems .
seen in their original context , the issues are usually more subtle . my aim is to show what can go wrong if techniques are used outside their original context .
over the past 60 years , the speed at which computers perform elementary calculations has increased by a factor 10@xmath0 , and the size of computer memories and the capacity of data storage devices have undergone similarly spectacular increases .
the earliest computer simulations of systems consisting of a few hundred atoms could only be performed on the world s largest computers .
now , anybody who has access to a standard computer for personal use can carry out simulations that would have required a supercomputer only 15 years ago .
moreover , software to carry out computer simulations is readily available .
the fact that the hardware and software thresholds for performing ` normal ' simulations have all but disappeared forces us to think about the role of computer simulations .
the key question is : why should one perform a simulation in the first place .
when we look at computer simulations in an applied context , the answer to the question ` why simulation ? ' is simple : they can save time ( and money ) .
increasingly , simulations are used to complement experiment or , more precisely , to guide experiments in such a way that they can focus on the promising compounds or materials .
this is the core of the rapidly growing field of computational materials science and computational ` molecular ' design .
computer simulations allow us to predict the properties of potentially useful substances , e.g. pharmaceutical compounds or materials with unique physical properties .
using computer simulations we can pre - screen candidate substances to minimise the amount of experimental work needed to find a substance that meets our requirements .
in addition , simulations are very useful to predict the properties of materials under conditions that are difficult to achieve in controlled experiments ( e.g. very high temperatures or pressures ) .
computational materials science of the type sketched above is the ` front end ' of a broader scientific endeavour that aims to advance the field of particle - based modelling , thus opening up new possibilities .
much of this development work is carried out in an academic environment where other criteria apply when we wish to answer the question whether a simulation serves a useful purpose .
below , i list several valid reasons to perform a simulation , but i also indicate what reasons i consider less convincing .
let me begin with the latter .
the total number of molecular systems that can , in principle , be simulated is very , very large .
hence , it is not difficult to find a system that nobody else has simulated before .
this may seem very tempting .
it is easy to perform a simulation , create a few nice colour snapshots and compute , say , a radial distribution function .
then , we write a manuscript for a high impact journal and , in the abstract , we write ` here , for the first time , we report molecular dynamics simulations of _
18-bromo-12-butyl-11-chloro-4,8-diethyl-5-hydroxy-15-methoxytricos-6,13-diene-19-yne-3,9-dione _ ' i took the name from wikipedia , and my guess is that nobody has simulated this substance .
then , in the opening sentence of our manuscript we write : ` recently , there has been much interest in the molecular dynamics of _ 18-bromo-12-butyl-11-chloro-4,8-diethyl-5-hydroxy-15-methoxytricos-6,13-diene-19-yne-3,9-dione ... _ ' and , with a few more sentences , a some snapshots and graphs , and a concluding section that mirrors the abstract , the work is done ... of course , this example is a parody of reality but only just .
such simulations provide information that answers no existing question it is like the famous passage in the hitchhikers guide to the galaxy , where the computer ` deep thought ' has completed a massive calculation to answer the question of _ life , the universe and everything_. the answer is 42 but the problem is that nobody really remembers what the question was .
a simulation should answer a question .
but there are different kinds of questions .
i will discuss some of the categories below .
our knowledge of forces between all but the simplest molecules is limited .
moreover , the construction of reliable force - fields is a time - consuming business that combines experimental and theoretical information .
the most widely used force fields are hopefully ` transferable ' .
this means that the interactions between molecules are decomposed into interactions between their constituent atoms or groups , and that the same interactions can be used when these atoms or groups are arranged to form other molecules
. the model may take interactions between charges and polarisability into account but , in the end , the force - field is always approximate .
this is even true when interactions are computed ` on the fly ' using density - functional theory or another quantum - based approach .
therefore , if we wish to apply a force field to a new class of molecules , or in a range of physical conditions for which it has not been tested , we can not assume _ a priori _ that our force - field will perform well : we must compare the predictions of our simulations with experiment .
if simulation and experiment disagree , our force field will have to be improved .
optimising and validating force fields is an important ingredient of computer simulations .
it has been suggested that simulation constitutes a third branch of science , complementing experiment and theory .
this is only partly true .
there is a two - way relation between theory and experiment : theories make predictions for experimental observations and , conversely , experiments can prove theories wrong
if experiment agrees with theory then this does not demonstrate that the theory is correct but that it is compatible with the observations .
simulations have some characteristics of theory and some of experiment . as is the case for theory , simulations start with the choice of a model , usually a choice for the hamiltonian ( ` the force field ' ) describing the system and , in the case of dynamical simulations , a choice for the dynamics ( newtonian , langevin , brownian etc ) . with this choice , simulations can ( in principle ) arrive at predictions of any desired accuracy the limiting factor is computing power
. however , simulations can never arrive at exact results , let alone analytical relations . unlike theories , simulations are therefore not a good tool to summarise our understanding of nature .
however , they can provide important insights : for instance , a simulation can show that a particular model captures the essential physics that is needed to reproduce a given phenomenon .
a case in point is hard - sphere freezing : before the simulations of alder and wainwright and jacobson and wood @xcite , it was not obvious that systems with hard - sphere interactions alone could freeze .
once this was known , approximate theories of freezing could focus on the simple hard - sphere model and ` computer experiments ' could be used to test these theories .
this shows two aspects of simulations : a ) it can act as a ` discovery tool ' and b ) it can be used to test predictions of approximate theories .
conversely , exact results from theory are an essential tool to test whether a particular algorithm works : if simulation results ( properly analysed and , if necessary , extrapolated to the thermodynamic limit ) disagree with an exact theoretical result , then there is something wrong with the simulation .
these tests need not be very sophisticated : they can be as simple as computing the average kinetic energy of a system and comparing with equipartition ( for a system of the same size ) , or computing the limiting behaviour of the non - ideal part of the pressure of a system with short - ranged interactions at low densities and comparing with the value expected on the basis of our knowledge of the second virial coefficient . when comparing simulation with experiment , we do something else : we are testing whether our hamiltonian ( ` force - field ' ) can reproduce the behaviour observed in experiment .
as said before : agreement does not imply that the force field is correct , just that it is ` good enough ' .
once a simulation is able to reproduce the experimental behaviour of a particular system , we can use the computer as a microscope : we can use it to find hints that help us identify the microscopic phenomena that are responsible for unexplained experimental behaviour .
experimentalists and simulators often speak a different ` language ' .
experimental data are usually ( although not always ) expressed in s.i . units .
simulators often use reduced units that can be an endless source of confusion for experimentalists ( see fig .
[ fig : reducedunits ] ) .
there are good reasons to use reduced units in simulations [ e.g. to use a characteristic particle diameter as a unit of length and to use the thermal energy ( @xmath1 ) as a unit of energy ] . however , the units should be clearly defined , such that a conversion to s.i .
units can easily be made .
there is one general rule , or rather a form of numerical ` etiquette ' , governing the comparison between experiment and simulation that is not often obeyed .
the rule is that , if at all possible , simulations should compute directly what is being measured in experiment .
the reason is that it is often not straightforward to convert experimental data into the quantities that are computed in a simulation : the procedure may involve additional approximations and , at the very least , is likely to lead to a loss of information . to give a specific example : scattering experiments do not measure the radial distribution function @xmath2 , rather , they measure @xmath3 over a finite range of wave - vectors . as @xmath2 is related to the fourier transform of @xmath3 , we would need to know @xmath3 for _ all _ wave vectors to compute @xmath2 . in practice ,
experiments probe @xmath3 over a finite window of wave - vectors .
if we would perform a fourier transform within this window , we would obtain an estimate for @xmath2 that contains unphysical oscillations ( and possibly even negative regions ) , due to the truncation of the fourier transform .
of course , procedures exist to mitigate the effects of such truncation errors , yet , if at all possible , simulations should report @xmath3 to save the experimentalist the trouble of having to perform an approximate fourier transform of their data .
there is , of course , a limit to this rule : some corrections to experimental data are specific for one particular piece of equipment .
if simulations would account for these specific instrument effects , they could only be compared to one particular experiment . in that case
, the meeting point should be half way .
computers have become more powerful but so have experimental techniques ( often for the same reason ) .
many experiments generate huge data streams that are impossible to analyse by hand . for this reason ,
computers play an important role in data analysis .
most of these techniques are not related to computer simulations but some are .
increasingly , molecular modelling is integrated in experiments .
the experimental data provide constraints that allow us to refine the model .
increasingly , this type of simulation will be embedded in the experimental equipment a black box that most users will not even see .
the staggering increase in computing power during the past 60 years [ roughly by a factor 10@xmath0 ] , is such that it has not only quantitatively changed computing , but also qualitatively , because we can now contemplate simulations ( e.g. a short simulation of a system the size of a bacterium ) that were not even on the horizon in the early 50 s .
however , in some cases , the speed - up achieved through the use of better algorithms ( e.g. techniques to study rare events ) have been even more dramatic . yet , the algorithms at the core of monte carlo or molecular dynamics simulations have barely changed since the moment that they were first implemented . hence , a computer simulator in the 60 s would have been able to predict fairly accurately on the basis of moore s law , what system sizes and time - scales we can simulate today .
however , he / she ( more ` he ' than ` she ' in the 50 s and 60 s arianna rosenbluth and mary - ann mansigh are notable exceptions ) would not have been able to predict the tools that we now use to study rare events , quantum systems or free - energy landscapes .
i should add that many algorithmic improvements are not primarily concerned with computing speed but with the fact that they enable new types of calculations ( think of thermostats , barostats , fluctuating box shapes etc ) .
it would be incorrect to measure the impact of these techniques in terms of a possible gain in computing speed . although this point of view is not universally accepted , scientists are human .
being human , they like to impress their peers . one way to impress your peers is to establish a record .
it is for this reason that , year after year , there have been and will be claims of the demonstration of ever larger prime numbers : at present
2012 the record - holding prime contains more than ten million digits but less than one hundred million digits . as the number of primes is infinite , that search will never end and any record is therefore likely to be overthrown in a relatively short time .
no eternal fame there . in simulations , we see a similar effort : the simulation of the ` largest ' system yet , or the simulation for the longest time yet ( it is necessarily ` either - or ' ) .
again , these records are short - lived .
they may be useful to advertise the power of a new computer , but their scientific impact is usually limited . at the same time , there are many problems that can not be solved with today s computers but that are tantalisingly close .
it may then be tempting to start a massive simulation to crack such a problem before the competition do it .
that is certainly a valid objective , and alder and wainwright s discovery of long - time tails in the velocity auto - correlation function of hard spheres is a case in point .
the question is : how much computer time ( ` real ' clock time ) should we be prepared to spend . to make this estimate ,
let us assume that the project will have unlimited access to the most powerful computer available at the time that the project starts .
let us assume that this computer would complete the desired calculation in @xmath4 years .
somewhat unrealistically , i assume that the computer , once chosen , is not upgraded .
the competition has decided to wait until a faster computer becomes available .
we assume that the speed of new computers follows moore s law .
then , after waiting for @xmath5 years , the time to completion on a new computer will be @xmath6 ( 2.17 years is the time it takes the computer power to increase by a factor of @xmath7 if moore s law holds ) .
the question is : who will win , the group that starts first , or the group that waits for a faster computer to become available : _ i.e. _ is @xmath8 the time @xmath5 that minimises @xmath9 , satisfies : @xmath10 this equation has a solution for positive @xmath5 when @xmath11 years . if the calculation takes less than 2.17 years , the group that starts first wins
otherwise , the group that waits finishes the calculation after @xmath12\ ] ] years .
of course , this calculation is very oversimplified , but the general message is clear : it is not useful to buy a computer to do a calculation that will take several years .
once more , i should stress that i mean ` wall clock ' time not processor years .
finally , it is important to realise that moore s law is not a law but an extrapolation that , at some point , will fail .
many of the subtleties of computer simulations are related to the fact that the systems that are being studied in a simulation are far from macroscopic .
the behaviour of a macroscopic system can be imitated , but never really reproduced , by using periodic boundary conditions .
the main advantage of periodic boundary conditions is that they make it possible to simulate a small system that is not terminated by a surface , as it is periodically repeated in all directions . but
even though the use of periodic boundary conditions will eliminate surface effects ( that are very dramatic for small systems ) , there are many other finite - size effects that can not be eliminated quite so easily .
below , i will give a few examples .
but first , a general comment about testing for finite size effects .
if a simulator suspects finite size effects in a simulation , the obvious test is to repeat the same simulation with a system of a larger size , e.g. a system with double the linear dimension of the original system .
the problem with this test is that it is rather expensive .
the computational cost of a simulation of a fixed duration scales ( at least ) linearly with the number of particles . for a 3d system
, doubling the linear dimensions of the system makes the simulation 8 times more expensive .
in contrast , halving the system size makes the simulation 8 times cheaper .
of course , if there is evidence for finite - size effects , then there it is necessary to perform a systematic study of the dependence of the simulation results and then there is no cheap solution .
a systematic analysis of finite size affects if of particular importance in the study of critical phenomena @xcite . in general ,
finite size effects will show whenever we study phenomena that are correlated over large distances : these can be static correlations ( as in the case of critical fluctuations ) or dynamic correlations , as in the case of hydrodynamic interactions .
when a particle moves in a fluid , it gradually imparts its original momentum to the surrounding medium . as momentum
is conserved , this momentum spreads in space , partly as a sound wave , partly as an ( overdamped ) shear waves .
the sound wave moves with the speed of sound @xmath13 and will therefore cross a simulation box with diameter @xmath14 in a time @xmath15 .
after that time , a particle will ` feel ' the effect of the sound waves emitted by its nearest periodic images . clearly , this is finite size effect that would not be observed in a bulk fluid .
the time @xmath16 for such spurious correlations to arise is , in practice , relatively short .
if @xmath14 is of the order of 10 nm and @xmath13 is of the order @xmath17 m / s , then @xmath16 is of order 10 ps .
needless to say that most modern simulation runs are much longer than that .
in addition to a sound wave , a moving particle also induces shear flow that carries away the remainder of its momentum .
this ` transverse ' momentum is transported diffusively .
the time to cross the periodic box is of order @xmath18 , where @xmath19 is the kinematic viscosity ( @xmath20 , where @xmath21 is the shear viscosity and @xmath22 the mass density ) . to take water as an example : @xmath23 and hence the time @xmath24 that it takes transverse momentum to diffuse one box diameter ( again ,
i will assume @xmath14 = 10 nm ) is around 10 ps .
what these examples show is that any numerical study of long - time dynamical correlations may require a careful study of hydrodynamic finite - size effects .
below , i will discuss the specific example of hydrodynamic interactions .
however , many of my comments apply ( suitably modified ) to the study of dislocations or the calculation of the properties of systems consisting of particles that interact through long - ranged forces ( e.g. charged or dipolar particles ) . at large distances ,
the hydrodynamic flow field that is induced by a force acting on a particle in a fluid , decays as @xmath25 .
if we now consider a system with periodic boundary conditions , the hydrodynamic flow fields due to all periodic images of the particle that is being dragged by the external force will have to be added .
it is clear that , just as in the case of the evaluation of coulomb interactions , a naive addition will not work .
it would even seem that the total flow field due to an infinite periodic array of dragged particles would diverge .
this is not the case , and the reason is the same a in the coulomb case : when we compute coulomb interactions in a periodic system , we must impose charge neutrality . similarly , in the case of hydrodynamic interaction , we must impose that the total momentum of the fluid is fixed ( in practice : zero ) .
this means that if we impart a momentum @xmath26 to a given article , then we must distribute a momentum @xmath27 among all other particles .
the result is that , if we start with a fluid at rest , dragging one particle in the @xmath28 direction , creates a back flow of all other particles in the @xmath29 direction .
if there are @xmath30 particles in the system , this is an @xmath31 effect .
but that is not the only finite size effect .
the other effect is due to the fact that the drag force per particle needed to move a periodic array of particles through a fluid is not simply the sum of the forces needed to move a single particle at infinite dilution .
in fact , the force is different ( larger ) by an amount that is proportional to @xmath32 ( roughly speaking , because the hydrodynamic interactions decay as @xmath25 ) .
hence , the mobility ( velocity divided by force ) of a particle that belongs to a periodic array is less than the of an isolated particle in a bulk fluid . as a consequence ( stokes - einstein ) ,
the diffusion coefficient of a particle in a periodic fluid has a substantial finite - size correction that scales as @xmath33 , where @xmath34 is the diameter of the particle .
for a three - dimensional system , the diffusion coefficient therefore approaches its infinite system - size limit as @xmath35 , _
i.e. _ very slowly . by comparison ,
the finite - size correction due to back flow is relatively minor .
periodic boundary conditions impose a specific symmetry on the system under study .
for instance , if we study a fluid using simple - cubic periodic boundary conditions , then a single ` cell ' of this periodic lattice may look totally disordered , yet as it is repeated periodically in all directions , the system should be viewed as a simple cubic crystal with a particularly complex unit cell . as was already well known in the early days of computer simulations , the imposed symmetry of the periodic boundary conditions induces some degree of bond - orientational order in the fluid @xcite . often , this ( slight ) order has no noticeable effect , but in the case of crystal nucleation it may strongly enhance the nucleation rate .
this is why finite size effects on crystal nucleation are so pronounced .
the shape of the simulation box has an effect on the magnitude of finite size effects . intuitively , it is obvious that this should be the case in the limit that one dimension of the box is much smaller than the others .
for instance , if the simulation box has the shape of a rectangular prism with sides @xmath36 , then the shortest distance between a particle and its periodic image is @xmath37 .
clearly , we want @xmath37 to be sufficiently large that a particle does not ` feel ' the direct effect of its periodic images .
this sets a lower limit for @xmath37 . for a cubic box
, @xmath38 and @xmath39 would have the same value , but for the rectangular prism mentioned above , they will be larger . hence , a large system volume would be needed to suppress finite size effect for a rectangular prismatic box than for a cubic box .
this is one of the reasons why cubic boxes are popular ; the other reason is that it is very easy to code the periodic boundary conditions for cubic boxes .
this second reason explains why cubic boxes are being used at all , because there are other box shapes that result in a larger distance between nearest periodic images for the same box volume .
depending on the criteria that one use , there are two shapes that can claim to be optimal .
the first is the _ rhombic dodecahedron _ ( i.e. the wigner - seitz cell of a face - centred cubic lattice ) . for a given volume , the rhombic dodecahedron has the largest distance between nearest periodic images or , what amounts to the same thing , it has the largest inscribed sphere .
the other ( more popular ) shape is the _ truncated octahedron _ ( the wigner - seitz cell of a body - centred cubic lattice ) . the truncated octahedron is the most ` spherical ' unit cell that can be used , in the sense that it has the smallest enclosing sphere .
in fact , the ratio of the radii of the enclosing and inscribed spheres of the truncated octahedron is @xmath40 . for a rhombic dodecahedron ,
this ratio is @xmath41 and for a cube it is @xmath42 . in two dimensions
the optimal box shape is a hexagon ( and in 4d it is the wigner - seitz cell of the so - called @xmath43 lattice ) . in simulations of isotropic fluids ,
any shape of the simulation box can be chosen , provided that it is sufficiently large to make finite - size effects unimportant . however , for ordered phases such as crystals , the choice of the simulation box is constrained by the fact that is has to accommodate an integral number of crystal unit cells . if the crystal under study is cubic , then a cubic simulation box will do , otherwise the simulation box should be chosen such that it is as ` spherical ' as possible , yet commensurate with the unit cell of the crystal under study . note
that this constraint implies that , if a given substance has several crystal polymorphs of lower than cubic symmetry , then different box shapes will be needed to simulate these phases .
moreover , for low symmetry crystals , the shape of the crystal unit cell will , in general , depend on temperature and pressure .
hence , simulations of different thermodynamic state points will require simulation boxes with different shapes .
if the shape of the simulation box is incompatible with the crystal symmetry , the crystal will be strained .
this is easy to verify by computing the stress tensor of the system : if the average stress is not isotropic ( hydrostatic ) , the shape of the simulation box causes a deformation of the crystal . as a consequence
, elastic free energy will be stored in the crystal and it will be less stable than in its undeformed state . not surprisingly , such a deformation will affect the location of phase coexistence curves . the effect of the shape of the simulation box on the crystal under study is not simply a finite size effect : even a macroscopic crystals can be deformed .
however , for a large enough system , it is always possible to choose the number of crystal unit cells in the @xmath44 and @xmath45 directions such that the resulting crystal is almost commensurate with the simulation box .
the remaining difference is a finite size effect .
however , if we have been less careful and have prepared a deformed crystal , then it will stay deformed , if not forever , then at least much longer than the time that we can study in a simulation . in practice ,
the shape of the simulation box is not chosen by trial and error until the crystal is stress free ( or , more precisely , has an isotropic stress ) . in the early 80 s , parrinello and
rahman @xcite developed a constant - stress molecular dynamics scheme that treated the parameters characterising the box shape as dynamical variables .
in a simulation , the box shape fluctuates and the average box shape is the one compatible with the imposed stress . if the imposed stress is isotropic , then the parrinello - rahman technique yields the box shape compatible with an undeformed crystal . shortly after its introduction
, the parrinello - rahman technique was extended to monte carlo simulations by najafabadi and yip @xcite .
more recently , the method of najafabadi and yip was used by filion and coworkers @xcite to predict the crystal structure of novel materials . in this method ,
small systems are used such that appreciable fluctuations in the shape of the simulation box are possible : this allows one to explore a wide range of possible structures .
potentially , the fluctuations in the shape of the simulation box can become a problem : if the box becomes very anisometric , the nearest - neighbour distance in some directions becomes very small and serious finite size effects are observed .
this problem is always present if fluctuating box shapes are used to simulated liquids ( but , usually , there is no need to use fluctuating box shapes in that case ) . in the case of crystals ,
the problems are most pronounced for small systems . in that case
, the fluctuations of the box shape may have to be constrained . to allow for larger changes in the shape of the crystal unit cell , the system should be mapped onto a new simulation box , once the original box becomes too deformed .
this remapping does not affect the atomic configuration of the system .
liquid crystals are phases with symmetry properties intermediate between those of an isotropic liquid and of a fully ordered , three - dimensional crystal
. the simplest liquid crystal , the so - called _ nematic _ phase , has no long - range translational order , but the molecules are on average aligned along a fixed direction , called the nematic _ director_. like isotropic liquids , nematic phases are compatible with any shape of the simulation box . however , for more highly ordered liquid - crystalline phases , the box shape should be chosen carefully . to give a simple example : the smectic - a phases is similar to the nematic phase in that all molecules are , on average aligned along a director but , unlike the nematic phase , the density of the nematic phase along the director is periodic . in the simplest case one can view the smectic - a phase as a stack of liquid - like molecular layers . clearly , in a simulation
, the boundary conditions should be such that an integral number of layers fits in the simulation box .
this constrains the system size in one direction ( viz . along the director ) , but not along the other two directions .
again , simulations with a flexible box shape can be used to let the system relax to its equilibrium layer spacing . however , as the smectic phase is liquid like in the transverse direction , one should make sure that the box shape in those directions is fixed .
there is a practical reason for using slightly more complex boundary conditions for smectic - a phases : the average areal density in all smectic - a layers is the same , but the number of molecules per layer will fluctuate .
this is a real effect .
however , the time it takes a molecule to diffuse ( ` permeate ' ) from one smectic layer to the next tends to be very long ( it involves crossing a free - energy barrier ) . hence ,
unless special precautions are taken , the distribution of molecules over the various smectic layers will hardly vary over the course of a simulation and hence , the system does not adequately sample the configuration space . this problem can be solved by using so - called ` helical ' boundary conditions .
suppose that we have a smectic phase in a rectangular prismatic simulation box .
we assume that the smectic layers are perpendicular to the @xmath45 axis and have an equilibrium spacing @xmath46 . now consider the next periodic box in the @xmath47-direction .
normally , this box is placed such that the point @xmath48 maps onto @xmath49 .
if we use helical boundary conditions , the box at @xmath37 is shifted by @xmath46 along the @xmath45 direction : hence , @xmath48 maps onto @xmath50 ( see fig . [
fig : helical ] ) .
the result is that all the smectic layers in the system are now connected laterally into a single layer .
now , fluctuations in the number of particles per layer is easy because it does not require permeation .
similar techniques can be used to allow faster equilibration of columns in the so - called _ columnar _ phase ( a phase that has crystalline order in two directions and liquid - like order in the third ) .
i should point out that helical boundary conditions are not necessarily more ` realistic ' than normal periodic boundary conditions .
in particular , the slow permeation of particles between layers is a real effect .
the simulation of cholesteric liquid crystals poses a special problem .
a cholesteric liquid crystal is similar to the nematic phase in that it has no long - ranged translational order .
however , whereas the director if a nematic phase has a fixed orientation throughout the system , the director of a cholesteric phase varies periodically ( ` twists ' ) in one direction and is constant in any plane perpendicular to the twist direction . for instance , if we choose the twist direction to be parallel to the @xmath45-axis , then the @xmath45-dependence of the director would be of the form @xmath51 where @xmath52 denotes the cholesteric pitch .
ideally , the periodic boundary conditions should be chosen such that the cholesteric pitch is not distorted . in practice , this is often not feasible because typical cholesteric pitches are very long compared to the molecular dimensions . for cholesteric phases of small molecules ( as opposed to colloids ) , pitches of hundreds of nanometers are common .
simulations of systems with such dimensions would be very expensive .
the problem can be alleviated a bit by using twisted periodic boundary conditions . in that case
the system is not simply repeated periodically in the @xmath45-direction ( the pitch axis ) , but it is rotated by an amount that is compatible with the boundary conditions @xcite . if the simulation box is a square prism , any rotation by a multiple of @xmath53 is allowed .
the minimal twist corresponds to the case where the rotation is @xmath54 . if the simulation box is a hexagonal prism , rotations that are a multiple of @xmath55 ( corresponding to 1/6-th of a pitch )
are allowed .
smaller angles are not possible . although @xmath56 , the box size needed to accommodate an undeformed cholesteric is usually still too large to be computationally feasible .
hence , cholesterics are usually simulated in an over- or under - twisted geometry and the properties of the undistorted are deduced using thermodynamics .
allen @xcite has used such an approach to compute the twisting power of chiral additives in a nematic phase . here
, i outline how a simulation of over- and under - twisted cholesterics can be used to estimate the equilibrium pitch .
to do so , we have to make use of the fact that , for small deformations , the free energy of a cholesteric phase depends quadratically on the degree of over(under)-twist : @xmath57 where @xmath58 and @xmath59 is the value of @xmath60 for which the system is torque - free and @xmath61 is the volume of the system .
the constant @xmath62 is the so - called ` twist elastic constant ' .
the subscript @xmath63 is there for historical reasons ( in a nematic , two other types of director deformation are possible : ` splay ' and ` bend ' : the associated elastic constants are denoted by @xmath64 and @xmath65 respectively ) . in what follows
, i drop the subscript of @xmath62 . in a periodic system with a box length @xmath39 in the @xmath45-direction , @xmath66 , where @xmath67 is the twist of the system over a distance @xmath39 . in practice
, we will only consider @xmath68 or @xmath69 .
then an infinitesimal change in the helmholtz free energy @xmath70 is given by @xmath71 where @xmath72 is the entropy of the system , @xmath73 the temperature , @xmath74 the pressure , @xmath75 the chemical potential ( per particle ) and @xmath30 the number of particles .
in what follows , i fix @xmath30 and @xmath73 .
it is useful to write @xmath76 as the sum of a volume change due to @xmath77 , a lateral expansion / compression of the system and @xmath78 , the volume change due to deformations in the @xmath45-direction .
note that the total free energy change due to a change in @xmath39 is given by @xmath79dl_z\;.\ ] ] we can write this as @xmath80dv_\parallel= -\left[p + k(\xi-\xi_0)\xi\right]dv_\parallel\;.\ ] ] a change in volume at constant pitch ( i.e. constant @xmath39 ) would yield @xmath81 the excess pressure for deformations along the @xmath45-axis is therefore @xmath82 by measuring @xmath83 we can determine @xmath59 .
momentum conservation in classical mechanics follows if the lagrangian of the system is invariant under infinitesimal translations .
this result carries over from a free system in the absence of external forces to a system with periodic boundary conditions . for a non - periodic system
, momentum conservation implies that the centre of mass of the system moves at a constant velocity . in the presence of periodic boundary conditions ,
this statement becomes ambiguous because one should specify how the position of the centre of mass is computed .
clearly if we interpret the centre of mass of the system as the mass - weighted average position of the particles in the simulation box at the origin , then the center of mass motion is not at all conserved .
every time the particle crosses the boundary of the simulation box and is put back at the other side the centre of mass undergoes a jump . however
, if we consider the center of mass of the particles that were originally in the ` central ' simulation box , and do not put these particles back in that box when they move to another box , then the velocity of the center of mass of those particles is conserved .
it should be stressed that the very concept of the ` boundary ' of the simulation box is confusing : for normal ( equilibrium ) simulations , we can draw the lattice of simulation boxes wherever we like , hence the boundaries between such boxes can be anywhere in the system . the ` boundary ' therefore has no physical meaning ( the situation is different in non - equilibrium simulations with so - called lees - edwards ` sliding ' boundary conditions ) . unlike linear momentum ,
angular momentum is not a conserved quantity in systems with periodic boundary conditions .
the basic reason is that the lagrangian is not invariant under infinitesimal rotations .
this invariance ( for spatially isotropic systems ) is at the origin of angular momentum conservation . in periodic systems ,
angular momentum is simply not uniquely defined . to see this ,
consider a system consisting of two particles zero net linear momentum . in a free system
, the angular momentum of this system is uniquely defined as it does not depend on the origin from which we measure the angular momentum .
however , it is easy to see that in a periodic system the apparent angular momentum will be different if we choose the origin of our coordinate system between the two particles in the same simulation box , or between a particle in one simulation box and the periodic image of the other particle . considering only particles in the same simulation box will not help because then the angular momentum will change discontinuously whenever we move a particle that has crossed a boundary back into the original simulation box .
there is another , more subtle , effect of periodic boundary conditions that is not always properly realised .
it has to do with the calculation of fourier transform of spatial correlation function . as an example ( and an important one at that ) , i consider the computation of the structure factor @xmath3 . for simplicity
, i will consider an atomic fluid .
@xmath3 is related to the mean square value of @xmath84 , the fourier component of the number density at wave vector @xmath85 : @xmath86 where @xmath87 note that the structure factor is , by construction , non - negative . in liquid state theory , we often use the fact that the structure factor of a fluid is related to the radial distribution function @xmath2 though @xmath88\ ; .\ ] ] in the thermodynamic limit , the volume of the system tends to infinity and the integral does not depend on the size of the integration volume .
however , in a small , periodic system it does .
more importantly , in an isotropic fluid , the radial distribution function is isotropic and we can write @xmath89\ ; .\ ] ] for a finite , periodic system , one might be tempted to replace the upper limit of this integration by half the box diameter . however ,
this is very dangerous because the function thus computed is no longer equivalent to the expression in eq .
( [ eq : rhoq2 ] ) .
in particular , the apparent @xmath3 is not guaranteed to be non - negative , even if one only considered those values of @xmath85 that are compatible with the periodic boundary conditions .
one solution to this problem is to extrapolate @xmath2 to larger distances , using a plausible approximation .
however , in general , it is safest to stick with eq .
( [ eq : rhoq2 ] ) , even though it is computationally more demanding .
computer simulations provide a powerful tool to locate first - order phase transitions . however , it is important to account properly for finite size effects .
the safest ( although not necessarily simplest ) way to locate first order phase transitions is to compute the free energy of the relevant phases in the vicinity of the coexistence curve . for a given temperature , this involves computing the free energy of both phases at one ( or a few ) state points and then computing the pressure of the system in the vicinity of these points . in this way
, one can evaluate the free - energy as a function of volume .
coexistence is then determined using a double tangent construction .
the advantage of this procedure is that finite - size effects are relatively small as the two phases are simulated under conditions where no interfaces are present .
however , the approach is rather different from the one followed in experiment where one typically looks for the point where the two phases are in direct contact .
if the interface between the two phases does not move , we have reached coexistence .
there are several reasons why , in simulations , this approach requires great care .
first of all , the creation of a two - phase system involves creating a slab of phase i in contact with a slab of phase ii ( the slabs are obviously periodically repeated ) .
the free energy of such a system is equal to the free energy of the coexisting bulk phases plus the total surface free energy .
a ( potential ) problem is that , if the slabs are not thick enough ( how thick depends on the range of the intermolecular potential and on the range of density correlations ) , the total surface free energy is not equal to that of two surfaces at infinite separation . a second problem
, that is more relevant for monte carlo than for molecular dynamics simulations is that two - phase systems equilibrate slowly .
in particular , coexistence between two phases requires that the pressures are equal . in molecular dynamics simulations ,
the rate of pressure equilibration is determined by the speed of sound typically such equilibration is rapid .
in contrast , in monte carlo simulations , pressure equilibrates through diffusion , and this is slower .
most likely , our first guess for the coexistence pressure at a given temperature will not be correct and we will see the phase boundary moving
. then we have to adjust the pressure .
however , it is very important that we do not impose an isotropic pressure on the system .
rather , we should vary the volume of the system by changing the box - length in the direction perpendicular to the interface .
the reason is that , in the transverse direction , the surface tension of the two interfaces contributes to the apparent pressure .
hence if the ` longitudinal ' pressure ( i.e. the one acting on a plane parallel to the interfaces ) is denoted by @xmath90 , then the apparent pressure in the perpendicular direction is @xmath91 in other words : the transverse pressure contains a contribution due to the laplace pressure of the interfaces .
the coexistence pressure is @xmath90 , not @xmath92 . in the above example
, we have considered the case of a liquid - liquid interface where the laplace correction is determined by the surface tension @xmath93 . however , when one of the coexisting phases is a crystalline solid the problems are more severe because the pressure inside a solid need not be isotropic . in that case , the procedure becomes more complicated .
first , one must determine the compressibility of the crystal or , more generally , the relation between the lattice constant of the bulk crystal and the applied isotropic pressure .
then the lateral dimensions of the crystalline slab are chosen such that it is commensurate with a crystal with the correct lattice constant for the applied longitudinal pressure . if the initial guess for the coexistence pressure was incorrect one should change the applied longitudinal pressure but also the lateral dimensions of the simulation box , such that the crystal again has the correct lattice constant corresponding to the applied pressure .
the rescaling of the lateral dimension of the simulation box should also be performed if we do not fix the longitudinal pressure but @xmath94 . in that case
we do not know a priori what the coexistence pressure will be , but we still must rescale the transverse dimensions of the simulation box to make sure that the stress in the bulk of the crystal is isotropic . finally , there is a general point that , for a small system , the free energy cost to create two interfaces is not negligible compared to the bulk free energy . for sufficiently narrow slabs the result may be that the system will form a strained homogeneous phase instead of two coexisting phases separated by interfaces
. in general , simulations that study phase equilibria by preparing two phases separated by interfaces will use biasing techniques to facilitate the formation of a two - phase system . for examples ,
see refs .
@xcite .
( [ eq : deltapgamma ] ) illustrates the surface tension of a liquid - liquid interface can create a difference between the longitudinal and transverse components of the pressure that can be measured in a simulation . in principle , eq .
( [ eq : deltapgamma ] ) can be used to determine the surface tension @xmath93 . however , eq .
( [ eq : deltapgamma ] ) is only valid for the interface between two disordered phases ( e.g. liquid - liquid or liquid - vapour ) .
if one of the phases involved is a crystal , the situation becomes more complicated .
in particular , the laplace pressure is not determined by the surface tension ( which , for a solid - liquid interface is called the ` surface free - energy density ' ) , but by the surface stress . to see why this is so , consider the expression of the surface free energy of an interface with area @xmath95 : @xmath96 the surface stress @xmath97 is the derivative of the surface free energy with respect to surface area : @xmath98 for a liquid - liquid interface , @xmath93 does not depend on the surface area @xmath95 , as the structure of the surface remains the same when the surface is stretched . hence , in that case @xmath99 .
however , for a solid , the structure of the surface is changed when the surface is stretched and hence @xmath93 changes . in that case ,
@xmath100 . as a consequence
it is much more difficult to determine @xmath93 for a solid - liquid interface than for a liquid - liquid interface .
in fact , whereas for a liquid - liquid interface @xmath93 can be obtained from eq .
( [ eq : deltapgamma ] ) , the determination of @xmath93 for a solid - liquid interface requires a free - energy calculation that yields the reversible work needed to create an interface between two coexisting phases .
an example of such a calculation can be found in ref .
@xcite .
classical mechanics and classical statistical mechanics are at the basis of a large fraction of all molecular dynamics and monte carlo simulations .
this simple observation leads to a trivial conclusion : the results of purely classical simulations can _ never _ depend on the value of planck constant and , indeed , the expressions that are used to compute the internal energy , pressure , heat capacity or , for that matter , the viscosity of a system are functions of the classical coordinates and momenta only .
however , there seems to be an exception : the chemical potential , @xmath75 . when we perform a grand - canonical simulation , we fix @xmath75 , @xmath61 and @xmath73 and compute ( for instance ) the average density @xmath101 of the system .
@xmath101 , @xmath61 and @xmath73 do not depend on @xmath102 , but @xmath75 does . to be more precise , we can always write @xmath75 as : @xmath103 where @xmath104 is a purely classical quantity that depends on the interactions between molecules . however , @xmath105 depends on planck constant .
for instance , for an atomic system , @xmath106 where @xmath107 is the thermal de broglie wavelength : @xmath108 for a molecular system , the expression for @xmath105 is of a similar form : @xmath109 where @xmath110 is the part of the molecular partition function associated with the vibrations and rotations ( and , possibly , electronic excitations ) of an isolated molecule .
@xmath110 is a function of @xmath73 but its value depends on @xmath102 .
it would then seem that the results of classical simulations can depend on @xmath102 .
in fact , they do not .
first of all , if we consider a system at constant temperature , the factors involving @xmath102 just add a constant to @xmath75 .
hence , the value of @xmath102 will not affect the location of any phase transitions : at the transition , the chemical potentials in both phases must be equal , and this equality is maintained if all chemical potentials are shifted by a constant amount .
one might object that the density of a system itself depends on @xmath75 and therefore on @xmath102 .
this is true , but in reality it is not a physical dependence : a shift the absolute value of the chemical potential is not an observable property .
one should recall that the chemical potential describes the state of a system in contact with an infinite particle reservoir at a given temperature and density . in experiments ,
such a reservoir can be approximated by a large system , but in simulations it is most convenient to view this reservoir as consisting of the same atoms or molecules , but with all inter - molecular interaction switched off . to see this , consider two systems at temperature @xmath73 : a system with @xmath30 particles in volume @xmath61 and a reservoir with @xmath111 particles in volume @xmath112 .
we assume that @xmath113 .
the density in the reservoir @xmath114 .
the reservoir and the system can exchange particles .
we now assume that the particle in the reservoir do not interact . to be more precise : we assume that all _ inter_-molecular interaction have been switched off in the reservoir , but all _ intra_-molecular interactions are still present .
this ideal gas is our reference system .
it is straightforward to write down the expression for the partition function of the reservoir for a given number of particles ( say @xmath111 ) : @xmath115 the total partition function ( system plus reservoir ) is : @xmath116 note that all terms in the sum have the same number of factors @xmath117 and @xmath110 .
i take these out and define @xmath118 through @xmath119 note that @xmath118 does not depend on @xmath102 .
now , the probability to find @xmath30 particles in the system is : @xmath120 dividing numerator and denominator by @xmath121 , and using the fact that for @xmath113 , @xmath122 , we can write @xmath123 note that in this expression , the number of particles in the system is controlled by @xmath124 , the number density in the reservoir . @xmath102 has disappeared , as it should .
the concept of a reservoir that contains an ideal gas of molecules that have normal intra - molecular interactions is extremely useful in practice , in particular for flexible molecules . in a grand canonical simulation , we prepare thermally equilibrated conformations of the molecules that we wish to insert . we can then attempt to insert one of these particles into the system that already contains @xmath30 particles .
the probability of insertion is determined by : @xmath125 where @xmath126 measures the interaction of the added particle with the @xmath30 particles already present , but _ not _ the intramolecular interactions of the added particle . note ( again ) that the crucial control parameter is @xmath124 , the hypothetical but purely classical density in the ideal reservoir .
occasionally , the role of planck constant can be a bit more subtle .
we have assumed that the quantised intramolecular motions of the particles do not depend on the inter - molecular interactions . at high densities ,
this approximation may break down .
usually , this means that the distinction between inter and intra - molecular degrees of freedom becomes meaningless .
however , if we insist on keeping the distinction , then we must account for the shift of intra - molecular energy levels due to inter - molecular interactions . in that case ,
@xmath102 necessarily enters into the description . of course , the examples above are but a small subset of the many subtleties that one may run into when performing a simulation .
my reason for discussing this particular set is that 1 ) they are important and 2 ) they illustrate that one should never use a simulation programme as a ` black box ' .
in the previous section , i discussed aspects of simulations that are more tricky than they might at first seem to be . in the present section ,
i list a few examples of simulations techniques that can not be used because they are intrinsically flawed . in a monte carlo simulation
, we compute thermal averages of the type @xmath127 in this expression , @xmath95 stands for any ` mechanical ' property of the system ( e.g. the potential energy or the virial ) .
what we can not compute in this way are ` thermal ' properties , such as the free energy of the entropy .
let us take the free energy as an example .
the relation between the helmholtz free energy @xmath70 and the partition function @xmath128 of the system is @xmath129 in what follows , i will focus on the configurational part of the partition function @xmath130 , as the remaining contributions to @xmath128 can usually be computed analytically .
@xmath131 clearly , the integral on the righthand side is not a ratio of two integrals , as in eqn .
[ eq : aav ] .
however , we can write @xmath132 we can rewrite this as @xmath133 and hence it would seem that we _ can _ express the partition function in terms of s thermal average that can be samples . however , the method will not work ( except for an ideal gas , or for similarly trivial systems ) .
the reason is that the most important contributions are those for which @xmath134 is very large , whilst the boltzmann weight ( @xmath135 ) is very small - these parts of configuration space will therefore never be sampled . for systems with hard - core interactions ,
the situation is even more extreme : an important contribution to the average comes for parts of configuration space where @xmath136 and @xmath137 . in other words , there is no free lunch : @xmath130 ( and therefore @xmath70 ) can not be determined by normal monte carlo sampling .
that is why dedicated simulation techniques are needed to compute free energies .
the widom ` particle - insertion ' method is used extensively to compute ( excess ) chemical potentials of the components of a moderately dense fluid .
below i briefly show the derivation of widom s expression . however , an almost identical derivation leads to an expression that relates the chemical potential to the boltzmann factor associated with particle removal . that expression is useless and , in the limit of hard - core interactions , even wrong .
consider the definition of the chemical potential @xmath138 of a species @xmath139 . from thermodynamics
we know that @xmath75 can be defined as : @xmath140 where @xmath70 is the helmholtz free energy of a system of @xmath30 particles in a volume @xmath61 , at temperature @xmath73 . for convenience ,
i focus on a one - component system .
if we express the helmholtz free energy of an @xmath30-particle system in terms of the partition function @xmath141 , then it is obvious that for sufficiently large @xmath30 the chemical potential is given by : @xmath142 .
for instance , for an ideal gas of atoms , @xmath143\;.\ ] ] the excess chemical potential of a system is defined as @xmath144}{v\int d{\bf r}^{n}\exp\left[-\beta u({\bf r}^n)\right]}\right\}\;.\end{aligned}\ ] ] we now separate the potential energy of the @xmath145-particle system into the potential energy function of the @xmath30-particle system , @xmath146 , and the interaction energy of the @xmath145-th particle with the rest : @xmath147 . using this separation , we can write @xmath148 as : @xmath149\exp\left[-\beta\delta u_{n , n+1}\right]}{v\int d{\bf r}^{n}\exp\left[-\beta u({\bf r}^n)\right]}\right\ } \nonumber\\ & = & -k_bt\ln\left\{\frac{\int d{\bf r}_{n+1}\left\langle\exp(-\beta\delta u_{n , n+1})\right\rangle_n}{v}\right\}\;,\end{aligned}\ ] ] where @xmath150 denotes canonical ensemble averaging over the configuration space of the @xmath30-particle system . in other words ,
the excess chemical potential is related to the average of @xmath151 over all possible positions of particle @xmath152 . in a translationally invariant system , such as a liquid
, this average does not depend on the position of the addition particle , and we can simply write @xmath153 the widom method to determine the excess chemical potential is often called the particle - insertion method " because it relates the excess chemical potential to the average of the boltzmann factor @xmath154 , associated with the random insertion of an additional particle in a system where @xmath30 particles are already present . to arrive at the ( incorrect ) ` particle - removal ' expression we simply write @xmath155}{\int d{\bf r}^{n+1}\exp\left[-\beta u({\bf r}^n+1)\right]}\right\}\;.\end{aligned}\ ] ] following the same steps as above ,
we then find : @xmath156 if the potential energy of the system is bounded from above ( and , of course , from below ) then this expression is not wrong , but simply rather inefficient .
the reason why it is inefficient is the same as in the case of sampling partition functions : the most important contribution come from regions of configuration space that are rarely sampled .
if the potential energy function is not bounded from above ( as is the case for hard - core interaction ) , eq .
( [ eqn : widomremoval ] ) simply yields the wrong answer . having said that , particle - insertion and particle - removal can be fruitfully combined in the so - called ` overlapping - distribution ' method @xcite .
but even the particle insertion method may occasionally yield nonsensical results .
this happens when applying the method to a system at high densities . as an example , let us consider a crystalline material . during a simulation of a perfect crystal containing @xmath30 particles
, we can carry out random particle insertions and we can compute the average boltzmann factor associated with such insertions . the method appears to work , in the sense that it gives an answer .
however , that answer is _ not _ the excess chemical potential of atoms in a crystal lattice .
rather , it is the excess chemical potential of _ interstitial _ particles .
the reason why the method does not yield the expected answer is that in a real crystal , there are always some vacancies .
the vacancy concentration is low ( ranging from around 1:10@xmath157 at melting , to vanishingly values at low temperatures ) . yet , for particle insertion , these vacancies are absolutely crucial : the contribution to the average of @xmath158 of a single insertion in a vacancy far outweighs all the contributions due to insertion in interstitial positions . using a combination of particle insertion and particle removal
@xcite we can obtain information about the equilibrium concentration of vacancies and interstitials .
because there are so few vacancies , a naive grand - canonical simulation of crystal will be inefficient . as the vacancy concentration is low very large systems
are needed to create enough vacancies where particles can be added to the system . of course
, one can perform biased grand - canonical simulations to alleviate this problem , but i am unaware of any attempts to do so . because the concentration of vacancies is so low in most ( but not all @xcite ) crystals ,
one subtlety is often overlooked .
the point is that , because of the presence of vacancies , the number of particles is not the same as the number of lattice sites ( i assume , for simplicity , a one - component bravais lattice ) .
the free energy of the system is then a function of the number of particles _ and _ the number of lattice sites .
we can then write the variation of the free energy at constant temperature as @xmath159 where @xmath160 denotes the number of lattice sites and @xmath161 the corresponding ` chemical potential ' .
i use quotation marks here because @xmath161 is not a true chemical potential : in equilibrium , the number of lattice sites is such that the free energy is a minimum and hence @xmath161 must be zero . however , in a simulation where we constrain @xmath30 and @xmath160 to be the same , @xmath161 is definitely not zero and incorrect predictions of the phase diagram result if this is not taken into account .
there are only few examples where the analysis of the chemical potential associated with lattice sites has been taken into account correctly @xcite .
i end this paper with a very brief summary some ` urban legends ' in the simulation field .
i should apologise to the reader that what i express are my personal views not everybody will agree . in the beginning of this paper
i pointed out that in a number of specific cases the computational gains made by novel algorithms far outweigh the increase in computing power expressed by moore s law .
however , the number of such algorithms is very small .
this is certainly true if we avoid double counting : successful algorithms tend to be re - invented all the time ( with minor modifications ) and then presented by the unsuspecting authors as new ( for example , simpson s rule has recently been reinvented in a paper that has attracted over 150 citations @xcite ) .
although i have no accurate statistics , it seems a reasonable to assume that the number of new algorithms that is produced every year keeps on growing .
unfortunately , these ` new ' algorithms are often not compared at all with existing methods , or the comparison is performed such that the odds are stacked very much in favour of the new method .
emphatically , i am not arguing against new algorithms
quite the opposite is true and , in fact , over the past few decades we have seen absolutely amazing novel techniques being introduced in the area of computer simulation .
i am only warning against the ` not invented here ' syndrome . [
[ molecular - dynamics - simulations - predict - the - time - evolution - of - a - many - body - system ] ] molecular dynamics simulations predict the time evolution of a many - body system ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ this view is more common outside the field than inside .
however , the outside community is much larger than the tribe of simulators .
the reason why molecular dynamics does not predict the time evolution of a many - body system is that the dynamics of virtually all relevant many - body systems is chaotic .
this means that a tiny error of the phase - space trajectory of the system will grow exponentially in time , such that in a very short time the numerical trajectory will bear no resemblance to the ` true ' trajectory with the same initial conditions . using more accurate algorithms
can postpone this so - called lyapunov instability .
however , doubling the accuracy of the simulation simply shifts the onset by a time @xmath4 .
doubling it again , will gain you another time interval @xmath4 .
the point is that the computational effort required to suppress the lyapunov instability below a certain preset level scales exponentially with simulation time : in practice , no simulation worthy of the name is short enough to avoid the lyapunov instability .
it would seem that this is a serious problem : if molecular dynamics does not predict the time evolution of the system , then what is the use of the method ? the reason why molecular dynamics simulations are still widely used is that ( most likely ) the trajectories that are generated in a simulation ` post - dict ' rather than ` pre - dict ' a possible real trajectory of the system .
in particular , it can be shown that the widely used verlet algorithm generates trajectories that are a solution to the discretised lagrangian equations of motion .
this means that the algorithm generates a trajectory that starts at the same point as a real trajectory , ends at the same point as that trajectory and takes the same time to move from the starting point to the end point . what molecular dynamics generates is a good approximation to a _
possible _ trajectory , not to the trajectory with the same initial conditions .
the distinction between the two types of trajectories is illustrated in fig .
[ fig : lyapunov ] .
typically , the time between the beginning and end points of a trajectory is much longer than the time it takes for the lyapunov instability to appear .
therefore , the trajectory that we generate need not be close to any real trajectory .
however , if we consider the limit where the time - step goes to zero , then the discretised lagrangian approaches the full lagrangian and the discretised path ( presumably ) approaches the real path .
i should stress that this result holds for the verlet algorithm and , presumably , for a number of verlet - style algorithms .
however , it is not obvious ( to me ) that the molecular dynamics trajectories generated in an event - driven algorithm with hard collisions are necessarily close to a real trajectory for long times . if we can believe molecular dynamics simulations ( and i believe that we can in a ` statistical ' sense ) then there are several excellent reasons to use the technique .
the first is , obviously , that molecular dynamics probes the dynamics of the system under study and can therefore be used to compute transport coefficients , something that a monte carlo simulation can not do .
another practical reason for using molecular dynamics is that it can be ( and , more importantly , has been ) parallellized efficiently .
hence , for large systems , molecular dynamics is almost always the method of choice .
this rule - of - thumb is old and nowhere properly justified .
there are many examples where it appears that the optimal acceptance of monte carlo moves should be substantially lower ( certainly for hard core systems ) .
also , in the mathematical literature @xcite it is argued ( but in a different context ) that the optimal acceptance of metropolis - style monte carlo moves is probably closer to 25 % ( 23.4 % has been shown to be optimal for a specific but not a statistical mechanical example ) .
a very simple example can be used to illustrate this point .
consider a hard particle in an ideal gas .
the number density of the ideal gas is denoted by @xmath101 .
if we move the hard particle over a distance @xmath162 , it sweeps out a volume proportional to @xmath162 .
to make life easy , i will consider a one - dimensional example . in that case
the volume swept out by the hard particle is equal to @xmath162 ( i will assume that the particle itself is much larger than @xmath162 ) .
the probability that a monte carlo move over a distance @xmath162 will be accepted is given by the probability that there will be no ideal gas particles in the volume swept out by the particle : @xmath163 if the trial displacement is @xmath162 , then the average _ accepted _ displacement is @xmath164\times 0 $ ] and the corresponding mean - squared displacement is @xmath165 of course , this system is trivial because we know that if we sample all possible configurations , the large particle will be distributed uniformly in the volume occupied by the ideal gas . again , in this case , it is plausible that the most efficient monte carlo algorithm is simply the one that makes the large particle ` diffuse ' most rapidly through configuration space .
now consider that we generate trial moves such that @xmath162 is uniformly distributed between @xmath166 and @xmath167 , where @xmath168 denotes the magnitude of the maximum displacement .
the diffusion of the large particle will be fastest for that value of @xmath168 that yields the largest mean - squared displacement .
we can compute this mean - squared displacement analytically : @xmath169}{(\rho\delta_m)^3}\ ; .\ ] ] as can be seen in fig .
[ fig : acceptance ] , the maximum value of @xmath170 corresponds to an average acceptance @xmath171/(\rho\delta_m)$ ] that is around 28 % ( i.e. considerably less than 50% ) .
does it matter ?
probably not very much : the very fact that there is an ` optimal ' acceptance ratio means that the efficiency is fairy flat near this optimum .
still , 25 % is rather different form 50 % .
when in doubt , check it out .
this is a point that i already mentioned above but i repeat it here : in normal ( equilibrium ) monte carlo or molecular dynamics simulations , the origin of the periodic box can be chosen wherever we like and we can remap the periodic boxes at any time during our simulation .
hence , the ` boundary ' of the periodic box is not a physical divide .
again , as said before , this situation changes for some non - equilibrium situations , e.g. those that use lees - edwards sliding boundary conditions @xcite .
there is another situation where the boundary of the simulation box has ( a bit ) more meaning , namely in the simulation of ionic systems . due to the long - range nature of the coulomb forces ,
the properties of the system depend on the boundary conditions at infinity .
this is not very mysterious : if we have a flat capacitor with a homogeneous polarised medium between the plates , then the system experiences a depolarisation field , no matter how far away the plates are .
however , if the capacitor is short - circuited , there is no depolarisation field .
something similar happens in a simulation of a system of charged particles : we have to specify the boundary conditions ( or , more precisely , the dielectric permittivity ) at infinity . if we choose a finite dielectric constant at the boundaries ( ` finite ' in this context means : anything except infinite )
then the internal energy of the system that we simulate has a polarisation contribution that depends quadratically on the dipole moment of the simulation box .
if we study a dipolar fluid , than the dipole moment of the simulation box does not depend on our choice of the origin of the box .
however , if we study a fluid of charged particles , then the choice of the boundaries does matter .
again , it is nothing mysterious : some crystal faces of ionic crystals are charged ( even though the crystal is globally neutral ) .
the depolarisation field in this crystal will depend on where we choose the boundary ( at a positively or at a negatively charged layer ) , no matter how large the crystal .
if the potential energy @xmath172 of a system is known as a function of the particles coordinates @xmath173 , then it is meaningful to speak about _ the _ energy ` landscape ' of the system : the potential energy of an @xmath30 particle system in @xmath46 dimensions will have minima , saddle points and maxima in the @xmath174 dimensional space spanned by the cartesian coordinates of the particles .
in contrast , it is not meaningful to speak of _ the _ free energy landscape : there are many . in general ,
free energies appear when we wish to describe the state of the system by a smaller number of coordinates than @xmath174 .
of course , using a smaller number of coordinates to characterise the system will result in a loss of information .
its advantage is that it may yield more physical insight in particular if we use physically meaningful observables as our coordinates .
examples are : the total dipole moment of a system , the number of particles in a crystalline cluster or the radius of gyration of a polymer .
typically , we would like to know the probability to find the system within a certain range of these new variables .
this probability ( density ) is related to the original boltzmann distribution . in what follows
, i will consider the case of a single coordinate @xmath128 ( often referred to as the ` order parameter ' ) .
the generalisation to a set of different @xmath128 s is straightforward .
note that @xmath128 may be a complicated , and often non - linear function of the original particle coordinates @xmath173 .
the probability to find the system in a state between @xmath128 and @xmath175 is @xmath176 , where @xmath177 is given by @xmath178 } { z } \;,\ ] ] where @xmath130 , the configurational part of the partition function , is given by @xmath179 we now _ define _ the free energy ` landscape ' @xmath180 as @xmath181 which implies that @xmath182 hence , the @xmath180 plays the same role in the new coordinate system as @xmath146 in the original @xmath174 dimensional system . the free energy @xmath180 may exhibit maxima and minima and , in the case of higher dimensional free - energy landscapes , free - energy minima are separated by saddle points .
these minima and saddle points play an important role when we are interested in processes that change the value of @xmath128 , e.g. a phase transition , or the collapse of a polymer .
the natural assumption is that the rate of such a process will be proportional to the value of @xmath183 at the top of the free - energy barrier separating initial and final states .
however , whilst , for reasonable choose of @xmath128 , this statement is often approximately correct , it is misleading because the height of the free energy barrier depends on the choice of the coordinates . to see this ,
consider another coordinate @xmath184 that measures the progress of the same transformation from initial to final state . for simplicity ,
i assume that i have a one - dimensional free - energy landscape and that @xmath184 is a function of @xmath128 only : @xmath185 .
then it is easy to show that @xmath186 the derivative on the right - hand side is the jacobian associated with the coordinate transformation from @xmath128 to @xmath184 . if @xmath184 is a linear function of @xmath128 , then the jacobian is harmless : it simply adds a constant to the free energy and hence does not affect barrier heights .
however , in the more general case where @xmath184 is a non - linear function of @xmath128 , the choice of the functional form affects the magnitude of the barrier height .
let us consider an extreme ( and unphysical ) examples to illustrate this .
we choose : @xmath187 then @xmath188 in other words : with this choice of coordinates , the barrier has disappeared altogether . in general , the choice of different order parameters will not affect the free energy barrier as drastically as this , but the fact remains : the height of the free energy barrier depends on the choice of the order parameter(s ) @xmath128 . of course , this dependence of the free - energy barrier on the choice of the order parameter is only an artefact of our description .
it can not have physically observable consequences . and ,
indeed , it does not .
the observable quantity is the rate at which the initial state transforms into the final state ( folding rate , nucleation rate etc . ) and that rate is independent of the choice of @xmath128 .
it is instructive to consider a simple example : the diffusive crossing of a one - dimensional barrier of height @xmath189 and with a flat top of width @xmath190 @xmath191\ ; , \ ] ] where @xmath192 is the heaviside step function .
we assume that the diffusion constant at the top of the barrier has a constant value @xmath193 .
moreover , we assume that , initially , the density is @xmath194 on the reactant side , and zero on the product side .
then the initial , steady - state flux across the barrier is given by : @xmath195 now we apply our transformation eq .
( [ eq : jacobiantransformation ] ) . in the region of the barrier
, we then get : @xmath196 this transformation will completely eliminate the barrier .
however , it should not have an effect to the reactive flux .
indeed , it does not , as we can easily verify : the width of the barrier in the new coordinates is @xmath197 and the diffusion constant becomes @xmath198 combining all terms , we get : @xmath199 which is identical to the original flux , as it should be .
there is another problem with free energy barriers : the fact that a coordinate @xmath128 joins two minima in a free energy landscape , does not necessarily imply that a physical path exists . again , we consider a simple example : a two dimensional system of a particle in a rectangular box .
the box is cut obliquely by an infinitely thin , but infinitely high , energy barrier ( see fig . [
fig : nopath ] ) .
we now choose the @xmath47 coordinate as our order parameter @xmath128 .
the free energy is computed by integrating the boltzmann factor over @xmath201 at constant @xmath128 .
the fact that the boltzmann factor vanishes on one point along a line of constant @xmath128 makes no difference to the integral .
hence , @xmath180 is constant .
the path between @xmath95 ( on the left ) and @xmath200 ( on the right ) appears barrier free . yet
, clearly , there is no physical path from @xmath95 to @xmath200 . 1 .
the construction of the coordinates @xmath128 is a matter of choice that should be guided by physical insight or intuition 2 .
the height of free energy barriers depends on the choice of these coordinates 3 .
the rate of ` barrier - crossing ' processes does _ not _ depend on the choice of the coordinates 4 .
the fact that there is no free - energy barrier separating two states does not necessarily imply that there is an easy physical bath between these states . as the field of numerical simulation expands ,
new techniques are invented and old techniques are rediscovered .
but not all that is new is true .
there are several examples of methods that had been discredited in the past that have been given a new identity and thereby a new lease on life .
usually , these methods disappear after some time , only to re - emerge again later .
the fight against myths and misconceptions never ends ( including the fight against fallacies that i have unknowingly propagated in this paper ) .
i am grateful to wilson poon for suggesting the topic of this paper .
what i wrote is necessarily very incomplete , but i enjoyed writing it .
i thank patrick varilly for a critical reading of this manuscript and for his many helpful suggestions .
i should stress that all remaining errors and misconceptions are mine alone .
this work was supported by the european research council ( erc ) advanced grant 227758 , the wolfson merit award 2007/r3 of the royal society of london and the engineering and physical sciences research council ( epsrc ) programme grant ep / i001352/1 .
99 , _ j. chem .
_ , * 27 * ( 1957 ) 1208 . , _ j. chem
_ , * 27 * ( 1957 ) 1207 . ,
_ a guide to monte carlo simulations in statistical physics _ ,
second edition ( cambridge university press , cambridge ) 2005 . , _ j. stat .
_ , * 15 * ( 1976 ) 299 .
, _ j. appl .
_ , * 52 * ( 1981 ) 7182 . ,
_ scripta metall .
_ , * 17 * ( 1983 ) 1199 . , _ phys
_ , * 103 * ( 2009 ) 188302 . ,
_ , * 79 * ( 1993 ) 277 . , _ phys . rev .
* 47 * ( 1993 ) 4611 .
a _ , * 25 * ( 1982 ) 1699 .
b _ * 81 * ( 2010 ) 125416 . ,
_ , * 49 * ( 1983 ) 1121 . ,
_ j. comp .
_ , * 22 * ( 1976 ) 245 . ,
_ j. chem .
b _ , * 105 * ( 2001 ) 6722 . , _ a monte carlo method for chemical potential determination in single and multiple occupancy crystals _ , arxiv:1209.3228 ( 2012 ) . ,
natl . acad .
usa _ , * 109 * ( 2012 ) 17886 .
* 109 * ( 2012 ) 17728 .
_ , * 99 * ( 2007 ) 235702 . ,
_ diabetes care _ , * 17 * ( 1994 ) 152 .
, _ annals appl .
_ , * 7 * ( 1997 ) 110 . ,
_ j. phys .
c _ , * 5 * ( 1972 ) 1921 . | this paper discusses the monte carlo and molecular dynamics methods .
both methods are , in principle , simple .
however , simple does not mean risk - free .
in the literature , many of the pitfalls in the field are mentioned , but usually as a footnote and these footnotes are scattered over many papers .
the present paper focuses on the ` dark side ' of simulation : it is one big footnote .
i should stress that ` dark ' , in this context , has no negative moral implication .
it just means : under - exposed . |
evidences for the existence of dark matter are mainly coming from cosmological observations related to the physics of gravity .
these include the relic density of dark matter , anisotropies in the cosmic microwave background ( cmb ) , large scale structure of the universe , as well as the bullet clusters and the associated gravitational lensing effects .
while we still do not know what the nature of dark matter is , it is clear that there is no room to accommodate dark matter in the standard model ( sm ) of particle physics based on gauge invariance of @xmath1 and einstein - hilbert gravity theory based on general coordinate invariance . while it is plausible that the nature of dark matter may have a purely gravitational origin , theories that have been put forward thus far are not as convincing as those from the particle physics point of view .
in particular the relic density strongly suggests that dark matter may be a weakly interacting massive particle ( wimp ) .
if dark matter can indeed be related to weak scale physics , there may be hope for us to detect them in various underground experiments of direct detection as well as in space experiments using balloons , satellites , or space station of indirect detection .
furthermore , wimp dark matter might be produced directly at the large hadron collider ( lhc ) by manifesting itself as missing energy with a spectrum that may be discriminated from standard model background of neutrinos . in this paper
, we will focus on the simplest dark matter model @xcite which is based on adding a real singlet scalar field to the sm .
the communication between the scalar dark matter and the sm gauge bosons and fermions must then go through the sm higgs boson .
while there have been many studies for this simple model and its variants in the literature @xcite , we believe a global study of this model is still missing . in this work
, we will fill this gap .
we use the current experimental constraints of relic density from wmap @xcite , 225 live days data from direct experiment xenon100 @xcite , diffuse gamma - ray flux from indirect detection experiment of fermi - lat using the dwarf spheroidal satellite galaxies ( dsphs ) @xcite , and a higgs boson candidate with mass about 125 gev reported recently by the lhc @xcite to deduce the best fit parameters of the model .
the deduced parameters are used to predict various phenomenology of the model at the lhc , including production of the mono-@xmath0 jet , single charged lepton , and two charged leptons plus missing energies .
we also evaluate the muon anomalous magnetic dipole moment which is a two loop process in the model . for a global fitting based on effective operators approach , see our recent work in @xcite .
a similar global analysis for isospin violating dark matter is presented in @xcite . in the next section
, we will briefly review the scalar phantom model of dark matter . in section iii
, we present the global fitting for the relevant parameters of the model using the various experimental constraints described above . in section iv , we discuss collider phenomenology and the muon anomalous magnetic dipole moment of the model .
we conclude in section v. some analytical formulas of the matrix elements needed in our analysis as well as the expression for the muon anomalous magnetic dipole moment are collected in the appendix .
the simplest dark matter model ( sz ) @xcite ( dubbed scalar phantom by the authors in @xcite ) is obtained by adding one real singlet scalar @xmath2 in addition to the higgs doublet @xmath3 to the sm .
the scalar part of the lagrangian is given by @xmath4 a discrete @xmath5 symmetry of @xmath6 while keeping all sm fields unchanged has been imposed to eliminate the @xmath2 , @xmath7 , and @xmath8 terms . as a result it guarantees the stability of the @xmath2 particle and hence it may be a viable candidate for wimp ( weakly interacting massive particle ) dark matter .
note that the @xmath9 term in eq.([sz ] ) implies a contact interaction vertex among the scalar dark matter .
the virtue of this model is its simplicity .
indeed , it represents the simplest realization of a broad class of models , in which we could add any number of singlet scalar @xmath2 to the standard model , or the standard model augmented by a private higgs sector @xcite .
the analysis given here is in the spirit of seeing whether or not the simplest version of this kind of model could now be ruled out .
after electroweak symmetry breaking , @xmath3 develops a vacuum expectation value @xmath10 , where @xmath11 gev . after making the shift @xmath12 , the physical higgs field @xmath13 obtains a mass @xmath14 and the last term in eq.([sz ] ) becomes @xmath15 the first term on the right handed side of eq.([shift ] ) implies the dark matter @xmath2
also pick up an additional contribution of @xmath16 to its mass , thus @xmath17 .
we will assume @xmath18 is always positive so that the @xmath5 symmetry will never be broken , except perhaps due to black hole effects .
the second term in eq .
( [ shift ] ) tells us that the dark matter @xmath2 can communicate to the sm fields and self - interact with itself via a tree level higgs exchange , while the last term contributes to the relic density calculation from the process @xmath19 if kinematically allowed .
if kinematics permits , the second term also allows higgs boson to decay into a pair of @xmath2 , giving rise to the invisible higgs width .
implication of invisible higgs width in the higgs search at the lhc will be discussed further in the following sections .
there are a few theoretical restrictions on the model , including vacuum stability , unitarity , and triviality .
stability of the vacuum requires the scalar potential be bounded from below . at tree level
, we have @xmath20 tree level perturbative unitarity constraints can be deduced by considering the longitudinal gauge boson scatterings @xcite as well as all scalar - scalar scatterings @xcite @xmath21 analysis of the triviality of this model can be found in the literature @xcite .
self - interacting cold dark matter was proposed in @xcite to resolve some conflicts between actual observations and wimp theory which predicts overly dense cores in the center of galaxies and clusters and an overly large number of halos within the local group .
the spergel - steinhardt bound @xcite for collisional ( self - interacting ) dark matter , @xmath22 can be used to constrain the contact self - coupling @xmath23 of the scalar phantom ( as well as the coupling @xmath24 through higgs exchange ) .
we refer to previous works @xcite on this issue .
in this section , we consider the global constraints coming from wmap relic density @xcite , the xenon100 data @xcite , the fermi - lat upper limit of diffuse gamma - ray flux based on dsphs @xcite , the lhc higgs mass around 125 gev @xcite , and the upper limit of higgs invisible decay branching ratio @xcite .
the relevant parameters of the model are the dark matter mass @xmath25 , the coupling @xmath24 and the higgs mass @xmath26 , which the likelihood functions will depend on .
the publicly available software package micromegas version 2.4.5 @xcite is used to calculate the relic density and gamma - ray flux . in the calculation of the relic density
, we consider all of the following two body tree level processes @xmath27 the relic density from the wmap 7-year result @xcite is @xmath28 where @xmath29 is the density of the cold dark matter normalized to the critical density and @xmath30 is the hubble rate in unit of 100 km sec@xmath31 mpc@xmath31 . in the fitting
, we use the gaussian distribution for the wmap relic density in our likelihood function , @xmath32 with the @xmath33 defined as to stand for the scalar phantom dark matter as well as to define the chi - squared . ]
@xmath34 where the @xmath35 can be read off from the wmap experimental error @xcite .
we also assume a theoretical uncertainty @xmath36 to be 10% of the prediction in order to account for the discrepancy due to different methods being used to solve the boltzmann equation in different relic density computation packages .
our working assumption is that the wmap data on relic density constrains the current model without affecting other cosmological parameters in a significant way .
here we need to calculate the elastic cross section for @xmath37 where @xmath38 is a nucleus with @xmath39 protons and @xmath40 neutrons . for a higgs mass around 125 gev ,
it is heavy enough to be integrated out to give an effective local interaction between the dark matter and the quarks .
since the local velocity of the dark matter is about @xmath41 , non - relativistic reduction is appropriate .
for the present model there is no spin - dependent cross section for the above elastic scattering , because the dark matter is a scalar particle .
the spin - independent cross section at zero recoil energy can be obtained as @xmath42 where @xmath43 is the reduced mass for the @xmath44 system and @xmath45 with @xmath46 denotes the number of heavy quarks and @xmath47 or @xmath48 . the factors @xmath49 and @xmath50 are hadronic matrix elements , and we will use their default values given in micromegas @xcite . for direct detection
we choose the truncated maxwell velocity distribution , which is the default choice in micromegas . because xenon100 is a counting experiment , the best choice for the likelihood function is poisson distribution , @xmath51 the @xmath52 and @xmath53 are the number of background and observation events taken from xenon100 @xcite , respectively . in order to achieve the minimum of @xmath54 equals to zero , we normalized eq .
( [ eq : xenonlike ] ) by the factor @xmath55 . for simplicity
we do not take the background uncertainties into account in our analysis .
the signal @xmath56 equals @xmath57 in the nuclear recoil energy range of 6.6 - 30.5 ke@xmath58 , where @xmath59 is the detector efficiency once the experimental cuts are applied to the total number of unbiased events @xmath60 . in other words
, @xmath61 is the fraction of @xmath62 generated by the monte - carlo simulation which survives the various cuts taken in xenon100 . our approximation of the xenon100 likelihood map as described in the text .
the dashed blue line shows the 90.0% cl bound .
it approximately equals to the xenon100 90% cl exclusion contour , shown by the thick solid red line .
the thin solid black line , the dashed magenta line , and the cyan dotted line show our calculations of the 68.0% cl , the 99.0% cl , and 99.99% exclusion bound , respectively .
, width=384 ] we use the following approximate method to evaluate the efficiency @xmath59 as a function of @xmath25 .
first , from xenon100 data of @xmath63 and @xmath64 at 90%c.l .
one can obtain a reference @xmath65 using the likelihood function just mentioned .
then , taking the values of @xmath66 and @xmath67 along the 90% c.l .
line from fig . 3 in @xcite as input to micromegas
, we can obtain @xmath62 which is a function of @xmath66 and @xmath67 .
hence , assuming the efficiency @xmath59 depends only on @xmath66 and @xmath67 , it is simply given by the ratio @xmath68 .
below when we compute the number of events for our model , we will multiply it by this @xmath59 to obtain the corresponding number of signals .
for poisson distribution we evaluate the effective @xmath33 according to the following expression @xmath69 in order to justify our simple treatment of the likelihood function shown in eq .
( [ eq : xenonlike ] ) , we plot the 68% , 90% , 99% , and 99.99% c.l .
curves , and compare to the published xenon100
curve in fig .
( [ xenonlikemap ] ) .
it is clear that our 90% c.l .
curve is almost identical to the one from xenon100 .
therefore , we can safely use the likelihood function in eq .
( [ eq : xenonlike ] ) .
since the couplings of the scalar dark matter to the sm fermions must go through the higgs exchange in the sz model , the positron flux and the anti - proton flux are probably too small for indirect detection .
we will focus on gamma - ray flux in indirect detection . for continuum gamma - ray ,
the relevant processes in the model are @xmath70 , @xmath71 , @xmath72 , and @xmath73 channels .
the @xmath71 , @xmath72 , and @xmath73 will subsequently decay into quarks and charged leptons which can produce gamma - rays .
however , the @xmath73 final state will be discarded in our later numerical analysis since micromegas does not provide the pythia decay table of @xmath74 gamma - rays . furthermore , fermi - lat @xcite has neither the limit for @xmath73 nor @xmath72 final state .
nevertheless , the @xmath73 contribution to gamma rays is expected to be small , since the higgs decays into sm fermion pairs are suppressed by the fermion mass . when @xmath75 , the dominant contribution comes from @xmath76 , and the next largest contribution comes from the heaviest fermion pair that the @xmath77 can annihilate into . for lower dm mass
, the dominant process is the following annihilation @xmath78 where the neutral pion coming from quark fragmentation decays into two photons .
detection of one or more spectral lines would be the smoking gun signal for dark matter . the following annihilation processes @xmath79 can give rise to discrete gamma lines . in the sz model , only the following two processes @xmath80 are possible .
the @xmath81 final state is possible if @xmath82 .
the gamma - ray lines are located approximately at energies @xmath83 and @xmath84 for the two final states @xmath85 and @xmath81 , respectively , with corrections of order @xmath86 , which is minuscule .
since these processes are one - loop induced and therefore suppressed , we do not include them in the global fitting .
instead , we will compute these cross sections after the scan and compare with the fermi - lat limits @xcite .
the dsphs of our milky way are satellite systems without active star formation or detected gas content .
they are thus fainter and expected to be dominated by dm due to their own gravitational binding .
although the expected flux of gamma - rays is not as high as the galactic center , these dwarf galaxies may have a better signal - to - noise ratio .
currently , the most stringent upper limits on the dm annihilation cross sections in various channels are derived by fermi - lat collaboration using the new 24 months data set with the following two improvements on their analysis @xcite .
first , they performed a joint likelihood analysis to 10 satellite galaxies which can improve their statistical power .
second , they included the uncertainties in the dark matter distribution in these satellites entered in the astrophysical @xmath87 factor @xmath88 \;,\ ] ] which is the line - of - sight integral of the squared dm density , @xmath24 , toward an observational direction , @xmath89 , integrated over a sustained solid angle , @xmath90 .
the gamma - rays flux is then given by @xmath91 where @xmath92 is the velocity - averaged pair annihilation cross section and @xmath93 is the gamma - ray energy distribution per annihilation .
based on these two improvements , robust upper limits of 95% c.l . on the @xmath94 for the @xmath95 , @xmath96 , @xmath97 , and @xmath76 channels
are derived in @xcite .
we will use these constraints on the diffuse gamma - ray flux in our global fitting . since each limit was obtained by assuming the dominance by one single channel
, we can approximately reconstruct the upper limit suitable to our case by applying the same method as in sec .
4.1 of @xcite . in our analysis for dsphs
, we adopt our likelihood function as follows @xmath98 where @xmath99 is the complementary error function and the effective @xmath33 is the same as eq.([effchisq ] ) .
in addition , because the astrophysical @xmath87 factor is expected to have a 3% uncertainty and the hadronization / decay tables in either micromegas @xcite or darksusy @xcite have a factor of @xmath100 uncertainty , we can then include the theoretical uncertainties as @xmath101 , where @xmath102 is the 95% c.l .
of the reconstructed upper limit for our dm pair annihilation cross section . in order to force our scan to go to the @xmath103 gev region , as suggested by the recent lhc data @xcite , we use the gaussian likelihood function for the higgs mass with a central value of @xmath104 gev and an experimental uncertainty @xmath105 gev .
since the @xmath106 is an input , we do not introduce any theoretical error for the higgs mass .
therefore , the likelihood function for the higgs mass is @xmath107 it has been pointed out recently in @xcite that the monojet search at the lhc has strongly disfavored @xmath108 for higgs - portal dark matter model where @xmath109 is the total sm higgs width . the invisible width of sm higgs in the sz model is @xmath110 we note that the invisible decay mode is not dominant in most of dark matter mass range .
only when @xmath111 gev and @xmath112 , the invisible decay of higgs becomes significant .
hence , we implement the higgs invisible decay as a 0/1 hard cut . if @xmath111 gev , @xmath112 , and @xmath113 , we multiply @xmath114 by 0 , otherwise by 1 .
the profile likelihood of @xmath115 for the sz model by the global fitting using wmap relic density @xcite , xenon100 @xcite , dsphs @xcite and a 125 gev higgs @xcite with an invisible branching ratio less than 40% .
, width=384 ] engaging with multinest v2.7 @xcite with 10000 living points , a stop tolerance factor 0.001 , and an enlargement factor reduction parameter 0.5 , we perform a random scan in the three dimensional parameter space of @xmath66 , @xmath24 , and @xmath106 restricted in the following ranges @xmath116 \leq & 3.0\nonumber\\ -3.0 & \leq \;\;\ ; \log_{10}[\rho ] \;\;\ ; \leq & 0.0 \\ 114.0\textrm { gev}&\leq \;\;\ ; \ ; m_{h } \;\;\ ; \ ; \leq & 130.0\textrm { gev } \nonumber\end{aligned}\ ] ] the selected scan range of @xmath24 is much smaller than the theoretical limit @xmath117 because the wmap window is very small which only allows @xmath118 .
furthermore , in order to scan efficiently in the higgs resonance region and cover the low @xmath24 region , we use the log priors for @xmath66 and @xmath24 as specified in eq.([domain ] ) .
similar results are found for the case of negative @xmath24 and will not be shown here . after hitting the stop criteria ,
we collect total 440682 samples , and plot 68% and 95% profile likelihood confidence limit contours based on 138017 samples which are selected by nested sampling algorithm @xcite .
the 68% and 95% confidence limit means that the total likelihood is greater than @xmath119 and @xmath120 , respectively .
the total likelihood function for our global fitting will be taken as @xmath121 and the effective total @xmath122 is given by @xmath123 our analysis uses the method of maximum likelihood .
the likelihood function of each experiment is listed clearly in eq .
( [ eq : rd ] ) for relic density , in eq .
( [ eq : xenonlike ] ) for the xenon100 data , in eq .
( [ eq : ind ] ) for the gamma - ray data of fermi - lat , and in eq.([eq : higgs ] ) that for the higgs boson mass .
the joint likelihood is then the product of all these likelihood functions , as given in eq .
( [ eq : total ] ) .
the best fit " point in fig . [ sp_mx_rho_like ] presented below ( as well as in fig .
[ sp_mx_sigsip_like ] ) is the point in the parameter space such that the joint likelihood function is maximum there .
the @xmath124 and @xmath125 regions in these figures are the 1@xmath35 and 2@xmath35 deviations relative from the best fit " point .
the result of the profile likelihood projected on the @xmath115 plane is shown in fig .
( [ sp_mx_rho_like ] ) .
we can clearly see that there are two branches : the vertical branch at low @xmath25 region and the horizontal branch hooked around at @xmath126100 gev .
the shape of these two branches is mainly due to the relic density constraint .
however , xenon100 and dsphs also play a significant role at the junction of the two branches , @xmath127 and @xmath128 , where relatively large @xmath129 and @xmath130 can be easily produced .
furthermore , the hard cut due to the higgs invisible branching ratio can remove some of the parameter space points with @xmath131 and @xmath132 .
on the other hand , it is hard to satisfy our constraints in the region @xmath133 , because the @xmath33 in this region rises sharply due to the higgs boson mass and relic density constraints .
the vertical branch in the figure is mainly due to the higgs resonance effect , which can efficiently enhance the dark matter annihilation cross section when @xmath134 falls near @xmath26 .
hence , the coupling @xmath24 has to be small correspondingly , in order to be consistent with wmap data . on the other hand , when @xmath135 , the @xmath136 channel dominates the annihilation cross section @xcite .
therefore , we can see from the figure that in the 1 and 2 @xmath35 c.l .
bands of the horizontal branch the allowed @xmath24 is roughly proportional to @xmath137 ( see eq.([sigmavvv ] ) at the appendix ) . the profile likelihood of the spin - independent cross section @xmath138 for the sz model projected onto the @xmath25 axis .
the latest xenon100 limits @xcite are overlaid for comparison .
the projected xenon-1 t sensitivity is also shown .
, width=384 ] in fig .
[ sp_mx_sigsip_like ] , we show the profile likelihood on @xmath25 - @xmath139 panel against the experimental 90% c.l . upper limit from xenon100 .
clearly , the xenon100 data is only able to rule out @xmath140 .
current dm direct detection can not constrain most of the parameters . on the other hand
, the higgs resonance region and most of the horizontal band can be tested in the future by xenon-1 t ( see the dashed line in fig .
[ sp_mx_sigsip_like ] ) .
other than the higgs resonance region , the wmap constraint dominates the likelihood function as shown in fig .
( [ sp_mx_sigsip_like ] ) , and therefore the largest likelihood of xenon100 only occurs at @xmath141 .
nevertheless , it is easier to satisfy the relic density constraint in the higgs resonance region , and therefore the largest likelihood of xenon100 in the higgs resonance region occurs at @xmath142 such that @xmath143 , by fine - tuning @xmath25 , @xmath24 , and @xmath26 . as a consequence , the best fit of our scan appears in the higgs resonance region .
with the result of the likelihood determined , we can proceed to evaluate other observables as predictions for the model , including gamma - ray lines , collider signatures , and muon anomalous magnetic dipole moment .
the annihilation cross sections for the gamma - ray line from @xmath144 ( left ) and @xmath145 ( right ) .
, title="fig:",width=288 ] the annihilation cross sections for the gamma - ray line from @xmath144 ( left ) and @xmath145 ( right ) . ,
title="fig:",width=288 ] in fig.([sp_mx_sigaaz ] ) , we plot the cross sections for the gamma - ray line in the sz model versus the profile likelihood projected onto the @xmath25 axis .
the left panel is for @xmath144 while the right one is for @xmath145 .
the fermi - lat data @xcite associated with different halo profiles are also shown for comparisons .
it is clear to see that the prediction for the @xmath144 annihilation cross section allowed by the profile likelihood is well below the fermi - lat data while that of @xmath145 is even further below the fermi - lat data .
hopefully , future better measurements made by fermi - lat can put a dent in the allowed profile likelihood .
if the invisible mode of @xmath146 opens up , we should study its impact on higgs search at the lhc ; in particular its effect on the branching ratios of @xmath147 , @xmath148 and @xmath149 , which apparently show some excesses over the background . since the current cms and atlas data @xcite showed that the excesses seen in @xmath150 , @xmath151 , and @xmath149 channels are consistent with the expectation of the sm higgs boson of 125 gev , and @xmath149 decay modes are slightly below while the @xmath150 mode is somewhat higher than the sm predictions .
] we can not allow the invisible decay mode to be too large ; otherwise the visible mode would become inconsistent with the current data .
it is easy to show that the branching ratio for a visible mode would be its sm branching ratio multiplied by @xmath152 where @xmath153 is the invisible branching ratio defined earlier as @xmath154 . in our scan in the previous section , we had required the invisible branching ratio @xmath155 such that each visible mode is reduced by an amount less than 40% so as not to upset the current data .
if the dark matter mass @xmath156 , the higgs boson simply behaves like the sm higgs boson . from our scan result in fig .
[ sp_mx_rho_like ] , a few typical points which have the likelihood within 68% c.l .
( 1-@xmath35 band ) can be identified as follows : 1 .
point a : @xmath157 gev , @xmath158 , @xmath159 gev .
the invisible branching ratio is right at @xmath160 .
the significance of this point is that the higgs boson still has a large branching ratio into @xmath77 .
the collider signature that we will discuss below consists of a large missing energy .
this is the point with the maximum likelihood , shown by the star in fig .
[ sp_mx_rho_like ] .
point b : @xmath161 gev , @xmath162 , @xmath163 gev .
this point gives @xmath2 a mass close to @xmath164 and hence above the higgs decay threshold .
this is the point at the low end of the second branch .
point c : @xmath165 gev , @xmath166 , @xmath159 gev .
this point gives a heavy @xmath2 that is still consistent with direct detection limits .
the most common search modes so far for the dark matter are the monojet and monophoton plus missing energies . in this model , monojet or monophoton production
must go through the higgs boson @xmath13 , so that the only sizable production cross sections have to go via @xmath167 in which we can attach a gluon to the @xmath0 or @xmath168 leg , or attach a photon line to the @xmath0 leg . since the @xmath0-parton luminosity is
small and gluon - fusion is a loop process , the monojet or monophoton rate would be relatively small . a feynman diagram showing mono-@xmath0 jet production with missing energies.,width=288 ] since the dm candidate @xmath2 only couples to the sm particles via the higgs boson , the @xmath2 will preferably couple to the heaviest fermion . at hadronic colliders ,
one of the interesting processes is @xmath169 where the superscript @xmath170 on the higgs boson denotes that the higgs boson could be on- or off - shell depending on the mass of @xmath2 . obviously , it is dominated by on - shell higgs boson for @xmath171 .
if @xmath171 and @xmath26 is lighter than @xmath172 , the higgs boson will dominantly decay into a pair of @xmath2 .
the corresponding collider signature would be a mono-@xmath0 jet plus missing energies .
the other possible signatures would be the associated production of the higgs boson with a gauge boson @xmath173 or @xmath39 : @xmath174 the final state in this case would consist of a charged lepton or a pair of charged leptons plus missing energies .
we calculate the event rates of mono-@xmath0 jet , single charged lepton , and a pair of charged leptons plus missing energies at the lhc-7 , lhc-8 , and lhc-14 .
we impose the following selection cuts for the @xmath0 jet or charged leptons and the transverse missing energy @xmath175 the cross sections for the mono-@xmath0 jet , single or a pair of charged lepton plus missing energies are tabulated in table [ x - sec ] .
the largest cross section comes from mono-@xmath0 jet production .
however , when we apply the @xmath176 gev cut the cross section mono-@xmath0 goes down 50 times .
after further imposing the @xmath177-tagging , the event rate would only be handful .
another interesting signature is the single charged lepton plus missing energies .
counting both negatively- and positively - charged leptons the cross section could be as high as 16 fb at the lhc-8 . given the lhc-8 can accumulate 20 fb@xmath31 each experiment , it would be more than 300 events each experiment
. the @xmath178 production would give , on the other hand , two charged lepton plus missing energies with a few times smaller event rates .
.[x - sec ] cross sections for mono-@xmath0 jet , single charged lepton or a pair of charged leptons plus missing energies arise from higgs boson production followed by @xmath179 .
we used the point a ( @xmath157 gev , @xmath158 , and @xmath159 gev , @xmath180 ) .
the selection cuts are defined in eq .
( [ cuts ] ) . [ cols="<,^,^,^ " , ] note that for other typical points of the model , e.g. , points b and c , the invisible decay mode of the higgs boson is closed , and therefore the decay is similar to the sm higgs boson . the process in eq .
( [ b ] ) will then give rise to @xmath181 or @xmath182 final states , depending on the higgs boson mass .
the processes in eq .
( [ l ] ) will give one or two charged leptons plus either @xmath183 or @xmath151 .
the sm background for the mono-@xmath0 jet plus missing energy would be similar to the current monojet search in atlas @xcite and cms @xcite , but now with a @xmath177-tag on the monojet .
the largest background @xcite comes from @xmath184 and @xmath185 with minor contributions from @xmath186 , single top production , and qcd multijets when leptons or extra jets get missing down the beam . on the other hand ,
background events with single or double charged leptons plus large missing energy comes from @xmath187 or @xmath188 with minor contributions from @xmath186 and single top production .
precise estimations of these backgrounds are beyond the scope of the present paper .
two loop feynman diagram contributing to the muon anomalous magnetic dipole moment.,width=192 ] the experimental value of the muon anomalous moment @xmath189 is @xmath190 while the sm prediction is @xmath191 the @xmath192 discrepancy between the above experimental measurement and theoretical calculations based on using the @xmath193 annihilation cross section for the estimation of the hadronic correction @xcite @xmath194 could be a harbinger of various new physics beyond the sm .
the contribution to the muon anomalous magnetic dipole moment @xmath195 in the sz model first shows up at the two loop level ( see fig .
[ [ gminus2 ] ] ) .
detailed expressions can be found in the appendix . in table
( [ muonanomaly ] ) , we show the numerical results of @xmath195 for the three typical points a , b and c from our scan .
ccc + point a & point b & point c + @xmath196 & @xmath197 & @xmath198 for all the relevant parameter space , we have checked that the contribution is negative and many orders of magnitude below the current experimental sensitivity .
the simplest dark matter model is realized by adding a real scalar singlet to the standard model as was discussed quite some time ago in @xcite , long before the popular dark matter candidate of neutralino in mssm model took the central stage . in this work
, we use the most current experimental constraints of the relic density from the 7 year wmap data , latest xenon100 data , annihilation cross sections from fermi - lat based on 10 dwarf spheroidal satellite galaxies of the milky way , as well as the 125 gev standard model higgs candidate as discovered recently by the lhc , to pin down the profile likelihood for the parameters @xmath24 and @xmath25 of the model .
the collected points are then used to evaluate the cross sections for the gamma - ray lines from @xmath144 and @xmath81 and found that they are well under the current limits from fermi - lat data .
a small part of the allowed parameter space around @xmath199 gev barely touches the fermi - lat data with the einasto halo profile .
recently , an interesting analysis in ref.@xcite using the fermi - lat data suggests there could be a gamma - ray line around 130 gev that may be related to dark matter annihilation .
however , other authors @xcite suggest that astrophysical sources like the fermi - bubbles @xcite could also be responsible for this line signal .
the gamma - ray lines in this simplest dark matter model can not accommodate this line signal based on the profile likelihood determined by the global fitting with the experimental constraints mentioned above .
we also study the lhc signals of mono-@xmath0 jet , single charged lepton or a pair of charged leptons plus missing energies of the model .
the most interesting case is the single charged lepton plus missing energies which can arise from associated production of @xmath200 followed by @xmath201 and invisible decay of the higgs . with a luminosity of 20 fb@xmath31 for each experiment of atlas and cms at lhc-8
, we expect several hundreds of such events based on the point a. we also evaluate the muon anomalous magnetic dipole moment of the model and found that it is many orders of magnitude below the current experimental limit for all relevant parameter space .
more stringent constraints are expected for this simple model of dark matter as more data from the lhc , direct and indirect detection experiments become available in the near future .
in this appendix , we list the matrix elements and annihilation cross sections for all the two body processes needed in the calculations of the relic density and indirect detection .
let @xmath56 to be the center of mass energy given by @xmath202 where @xmath203 with @xmath204 being the velocity of the dark matter .
@xmath205 is the color factor , 1 for leptons and 3 for quarks .
\(1 ) @xmath206 : @xmath207 @xmath208 \(2 ) @xmath209 ( @xmath210 or @xmath39 ) : @xmath211 \label{eq : chichivv}\ ] ] @xmath212\ ] ] here @xmath213 is a kronecker delta to account for the bose statistics of the @xmath72 final state .
we note that micromegas computes process cross sections by calchep @xcite .
however , we found that the amplitude squared of @xmath214 differs between eq .
( [ eq : chichivv ] ) and the result from calchep .
the factor inside the square bracket of eq .
( [ eq : chichivv ] ) is @xmath215 $ ] , while the corresponding factor in calchep reads @xmath216 $ ] . due to this discrepancy
we rescale the cross section by the ratio of these two factors .
\(3 ) @xmath217 @xmath218 @xmath219
\(4 ) @xmath145 @xmath220 @xmath221 @xmath222 } { m_w^2 - y(1-y)m_z^2 + xy ( m_z^2 - s ) - i 0^+ } \nonumber \\ & = & ( 3 - \tan^2 \theta_w ) i \left ( \tau_w , \tau_{zw } \right ) + \left [ ( - 5 + \tan^2 \theta_w ) - 2 ( 1 - \tan^2 \theta_w ) \tau_w \right ] j(\tau_w , \tau_{zw})\end{aligned}\ ] ] @xmath223 here , @xmath224 , @xmath225 , @xmath226 , and @xmath227 . @xmath228 and @xmath87 are given by @xmath229 @xmath230 \nonumber \\ & & + \frac{1}{4 } \tau_{2 } \left ( \tau_1 - \tau_2 \right)^{-2 } \left ( g\left(\tau_1\right ) - g\left ( \tau_2 \right ) \right)\end{aligned}\ ] ] with @xmath231 and @xmath168 defined by @xmath232 ^ 2 & \;\ ; \mbox{for } \tau \leq 1 \\
-\frac{1}{4 } \left [ \ln \left(\frac{1 + \sqrt{1 - \tau^{-1}}}{1 - \sqrt{1 - \tau^{-1 } } } \right ) - i \pi\right]^2 & \;\ ; \mbox{for } \tau > 1 \end{cases}\ ] ] @xmath233 @xmath234 ( 5 ) @xmath144 @xmath235 @xmath236 @xmath237 @xmath238\ ] ] with @xmath231 defined in eq.([f ] ) .
@xmath239 ( 6 ) @xmath240 @xmath241 @xmath242 where @xmath243 is given by eq.([a_f ] ) .
@xmath244 following a similar procedure as in the qed case @xcite , one can readily obtain the following result for the muon anomalous magnetic dipole moment @xmath245 ^ 4 } \int_0 ^ 1 dz \log \left [ 1 + \chi(z ) \xi^2 \right]\ ] ] where @xmath246 with @xmath247 and @xmath248 . performing the @xmath249 integral
, we end up with a two - dimensional integration for @xmath195 @xmath250 where @xmath251\ ] ] and @xmath252
this work was supported in parts by the national science council of taiwan under grant nos .
99 - 2112-m-007 - 005-my3 and 101 - 2112-m-001 - 005-my3 as well as the wcu program through the kosef funded by the mest ( r31 - 2008 - 000 - 10057 - 0 ) .
az was supported by the nsf under grant no .
phy07 - 57035 ; he is also grateful to the institute of physics of the academia sinica , taiwan , for its warm hospitality .
tcy is grateful to ncts and kitpc for their warm hospitalities .
yst was funded in part by the welcome programme of the foundation for polish science .
c. p. burgess , m. pospelov and t. ter veldhuis , nucl .
b * 619 * , 709 ( 2001 ) [ hep - ph/0011335 ] .
y. cai , x. -g .
he and b. ren , phys .
d * 83 * , 083524 ( 2011 ) [ arxiv:1102.1522 [ hep - ph ] ] .
m. gonderinger , y. li , h. patel and m. j. ramsey - musolf , jhep * 1001 * , 053 ( 2010 ) [ arxiv:0910.3167 [ hep - ph ] ] .
h. davoudiasl , r. kitano , t. li and h. murayama , phys .
b * 609 * , 117 ( 2005 ) [ hep - ph/0405097 ] .
a. drozd , b. grzadkowski and j. wudka , jhep * 1204 * , 006 ( 2012 ) [ arxiv:1112.2582 [ hep - ph ] ] .
e. komatsu _ et al .
_ [ wmap collaboration ] , astrophys .
* 192 * , 18 ( 2011 ) [ arxiv:1001.4538 [ astro-ph.co ] ] g. aad _ et al . _ [ the atlas collaboration ] , arxiv:1207.7214 [ hep - ex ] .
s. chatrchyan _ et al . _
[ the cms collaboration ] , arxiv:1207.7235 [ hep - ex ] .
k. cheung , p. -y .
tseng , y. -l .
s. tsai and t. -c .
yuan , jcap
* 1205 * , 001 ( 2012 ) [ arxiv:1201.3402 [ hep - ph ] ] . h. -b .
jin , s. miao and y. -f .
zhou , arxiv:1207.4408 [ hep - ph ] .
r. porto and a. zee , phys .
b666 * , 491 , 2008 [ arxiv:0712.0448v3 [ hep - ph ] ] ; phys .
d * 79 * , 013003 , 2009 [ arxiv : 0807.0612 [ hep - ph ] ] ; y. bentov and a. zee , arxiv : 1207.0467 [ hep - ph ] .
b. w. lee , c. quigg and h. b. thacker , phys .
d * 16 * , 1519 ( 1977 ) .
d. n. spergel and p. j. steinhardt , phys .
lett . * 84 * , 3760 ( 2000 ) [ astro - ph/9909386 ] .
d. e. holz and a. zee , phys .
* b517 * , 239 - 242 ( 2001 ) [ hep - ph/0105284 ] .
a. djouadi , a. falkowski , y. mambrini and j. quevillon , arxiv:1205.3169 [ hep - ph ] ; a. djouadi , o. lebedev , y. mambrini and j. quevillon , phys .
b * 709 * , 65 ( 2012 ) [ arxiv:1112.3299 [ hep - ph ] ] .
g. belanger , f. boudjema , p. brun , a. pukhov , s. rosier - lees , p. salati and a. semenov , comput .
commun . * 182 * , 842 ( 2011 ) [ arxiv:1004.1092 [ hep - ph ] ] .
a. goudelis , y. mambrini and c. yaguna , jcap * 0912 * ( 2009 ) 008 [ arxiv:0909.2799 [ hep - ph ] ] . c. e. yaguna , jcap * 0903 * , 003 ( 2009 ) [ arxiv:0810.4267 [ hep - ph ] ] .
m. ackermann _ et al .
_ [ lat collaboration ] , arxiv:1205.2739 [ astro-ph.he ] .
l. roszkowski , e. m. sessolo and y. -l .
s. tsai , arxiv:1202.1503 [ hep - ph ] .
p. gondolo , j. edsjo , p. ullio , l. bergstrom , m. schelke and e. a. baltz , jcap * 0407 * , 008 ( 2004 ) [ arxiv : astro - ph/0406204 ] .
f. feroz , m. p. hobson and m. bridges , mon . not .
soc . * 398 * , 1601 ( 2009 ) [ arxiv:0809.3437 [ astro - ph ] ] .
f. feroz and m. p. hobson , mon . not .
soc . * 384 * , 449 ( 2008 ) [ arxiv:0704.3704 [ astro - ph ] ] .
x. -g .
he , s. -y .
ho , j. tandean and h. -c .
tsai , phys .
d * 82 * , 035016 ( 2010 ) [ arxiv:1004.3464 [ hep - ph ] ] .
g. aad _ et al .
_ [ atlas collaboration ] , phys .
b * 705 * , 294 ( 2011 ) [ arxiv:1106.5327 [ hep - ex ] ] .
s. a. malik , arxiv:1110.1609 [ hep - ex ] .
particle data group , review of particle physics , journal of physics g , nuclear and particle physics , vol . *
37 * , no 7a 075021 ( 2010 ) and 2011 partial update for 2012 edition . c. weniger , arxiv:1204.2797 [ hep - ph ]
s. profumo and t. linden , arxiv:1204.6047 [ astro-ph.he ] . m. su , t. r. slatyer and d. p. finkbeiner , astrophys .
j. * 724 * , 1044 ( 2010 ) [ arxiv:1005.5480 [ astro-ph.he ] ] .
a. belyaev , n. d. christensen and a. pukhov , arxiv:1207.6082 [ hep - ph ] .
s. weinberg , the quantum theory of fields , volume i , cambridge university press ( 1995 ) . | we present a global study of the simplest scalar phantom dark matter model .
the best fit parameters of the model are determined by simultaneously imposing ( i ) relic density constraint from wmap , ( ii ) 225 live days data from direct experiment xenon100 , ( iii ) upper limit of gamma - ray flux from fermi - lat indirect detection based on dwarf spheroidal satellite galaxies , and ( iv ) the higgs boson candidate with a mass about 125 gev and its invisible branching ratio no larger than 40% if the decay of the higgs boson into a pair of dark matter is kinematically allowed .
the allowed parameter space is then used to predict annihilation cross sections for gamma - ray lines , event rates for three processes mono-@xmath0 jet , single charged lepton and two charged leptons plus missing energies at the large hadron collider , as well as to evaluate the muon anomalous magnetic dipole moment for the model . |
cataclysmic variables ( cvs ) are close binary systems in which a late - type star ( secondary ) fills its roche lobe and transfers mass through the inner lagrangian point to a white dwarf ( primary ) .
gw librae was the first cv discovered with an accreting white dwarf showing photometric variations consistent with nonradial g - mode pulsations observed in non - interacting white dwarf stars @xcite . this discovery has opened a new venue of opportunity for us to learn about the stellar parameters of accreting variable white dwarfs using asteroseismic techniques .
we can probe the insides of a white dwarf using asteroseismology just like seismologists on earth can learn about the interior of our planet using earthquakes .
a unique model fit to the observed periods of the variable white dwarf can reveal information about the stellar mass , core composition , age , rotation rate , magnetic field strength , and distance ( see the review paper * ? ? ?
this information could prove vital in determining the system dimensions and parameters , and may be the only way to derive conclusive results for non - eclipsing cvs .
pulsating white dwarfs in cvs are also interesting to the variable white dwarf community because they can help establish the effect of accretion on pulsations .
we report new accreting pulsators , sdssj074531.91 + 453829.5 and sdssj091945.10 + 085710.0 ( hereafter sdss0745 + 4538 & sdss0919 + 0857 ) , originally discovered to be cvs in the sloan digital sky survey ( sdss ) by @xcite and @xcite respectively .
this result comes from our long - term search to discover additional accreting pulsating white dwarf stars in cvs .
there are now eleven accreting pulsating white dwarfs known , including the two presented in this paper ( see * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
our goal is to establish the pulsationally unstable region(s ) for accreting white dwarfs , which requires a statistically significant number of these systems with well - determined temperatures .
finding the non - variables around an instability strip is equally important in determining its boundaries as discovering variables close to its edges . with this context in mind
, we also report a non - variable sdssj171145.08 + 301320.0 ( hereafter sdss1711 + 3013 ) , discovered as a cv by @xcite .
we observed the system twice and place useful non - variability limits related to the absence of pulsation in the primary white dwarf .
we select those cvs for high - speed photometry whose optical spectra include prominent broad absorption lines from the white dwarf as well as the contaminating narrow emission features from the accretion disk .
the absorption lines indicate that the white dwarf flux dominates the light from the cv .
while this is not a _ sufficient _ criterion for pulsation in the primary white dwarf , these are the only systems where the search is even possible .
when the orbital period of a cv is @xmath180 - 90min , it is near the evolutionary period minimum and the cv exhibits the smallest rate of mass transfer .
@xcite used the optical accretion luminosity to constrain the rate of mass transfer to be greater than @xmath2 , while @xcite and @xcite theoretically compute the rate of mass transfer at the period minimum to be few times @xmath3 .
we can expect that the low rates of mass transfer allow a larger surface area of the white dwarf to be visible , making cvs near the period minimum ideal candidates to find additional accreting pulsators .
note that the two criteria mentioned here are not independent .
@xcite investigate the temperature range in which models of accreting white dwarfs with a wide range of masses and helium enrichment from the donor star would be pulsationally unstable .
they find a h / hei instability strip for accreting model white dwarfs with a blue edge near @xmath412000k for a 0.6@xmath5 star .
the blue edge shifts to hotter ( cooler ) temperatures by about 2000k for a factor of 10 increase ( decrease ) in gravity ; we can expect the blue edge at 14000k corresponding to @xmath6=9 and at 10000k for @xmath6=7 .
this theoretical instability strip is similar to the zz ceti instability strip @xcite , and are also known as the zz ceti stars . ] . for accreting model white dwarfs with a high he abundance ( @xmath70.38 ) , @xcite find an additional hotter instability strip at @xmath815000k due to heii ionization .
the boundaries of this intermediate instability strip depend on the helium abundance and the mass of the accreting white dwarf . for a he abundance higher than 0.48 ,
these theoretical instability strips essentially merge .
@xcite expect that there are thus two distinct instability strips for accreting model white dwarfs with a he abundance between about 0.38 and 0.48 .
optical spectra do not lead to sufficiently reliable temperatures for the white dwarf due to contamination from the accretion disk .
accurate temperature determination requires ultraviolet spectra ( see * ? ? ?
* ) that we do not have .
hence we are unable to impose any additional selection criteria that depend on the temperature of the primary white dwarf , despite the theoretical framework mentioned above .
besides , an empirical instability strip for accreting white dwarfs has not yet been established .
there are only five accreting pulsators with well determined temperatures from ultra - violet spectra ; gw librae , sdss013132.39 - 090122.3 , sdssj161033.64 - 010223.3 , and sdssj220553.98 + 115553.7 lie within the range 1450015000k @xcite , while @xcite find the accreting white dwarf pulsator hs2331 + 3905 to be at 10500k .
although we do not use temperatures from optical spectra to select candidates , we estimated temperatures for the two pulsators and the non - pulsator discovered during our search . in order to establish the white dwarf temperature , the distance to the system , and to constrain the spectral type of the donor star , we model the optical sdss spectra as the sum of a white dwarf , an accretion disk , and a late - type secondary star . for the white dwarf
, we assume a surface gravity of @xmath9 , corresponding to a mass of @xmath10 and a radius of @xmath11 cm .
the contaminating emission from the accretion disk fills in the balmer absorption from the white dwarf , making it impossible to determine @xmath12 g independently .
we then compute a grid of pure hydrogen model spectra covering effective temperatures in the range 800020000k using the codes tlusty / synspec @xcite .
we model the accretion disk as an isothermal / isobaric slab of hydrogen @xcite . for the donor star
, we use the m - dwarf templates of @xcite and l - dwarf templates from @xcite .
we fix the radius of the secondary star to @xmath13 cm , which corresponds to a stellar mass of @xmath14 , close to the dividing line between main - sequence and sub - stellar objects , at an orbital period of 80min .
the free parameters of our three - component model are the white dwarf temperature @xmath15 , the distance @xmath16 , the spectral type of the secondary star sp(2 ) , the disc temperature @xmath17 , and its surface density @xmath18 . considering the number of free parameters , we refrain from applying a formal @xmath19 fit and model the sdss spectrum as follows .
we initially calculate a disk spectrum for the given choice of @xmath17 and @xmath18 , and scale the h@xmath20 emission line fluxes of the model to the observed value .
next , for the chosen @xmath15 , we adjust @xmath16 to reproduce the observed flux , and judge the goodness of the fit by eye .
we ensure that the combination of the white dwarf and the disk fits the balmer absorption lines in the sdss spectrum , as well as the slope of the continuum at @xmath21 , where the contribution of the donor star can safely be ignored .
finally , we scale the selected m / l dwarf template for @xmath16 as determined from the white dwarf fit .
the absence of noticeable tio bands in the sdss spectrum provides an _ early _ limit on the spectral type of the donor _ later _ spectral types can not be excluded .
we list the parameters corresponding to our model fits to the sdss spectra in table[t - fitparameter ] , and show these models along with the observed spectra in figure[t - fit ] .
the white dwarf temperatures estimated from the optical spectra are uncertain by @xmath2210002000k .
llllll sdss0745 + 4538 & 11000 & @xmath7m8 & 280 & 6600 & @xmath23 + sdss0919 + 0857 & 13000 & @xmath7m9 & 235 & 6300 & @xmath24 + sdss1711 +
3013 & 10500 & @xmath7m8 & 480 & 6500 & @xmath25 + we show the observed optical spectra of sdss0745 + 4538 ( top panel ) , sdss0919 + 0857 ( middle panel ) , and sdss1711 + 3013 ( bottom panel ) in gray along with their respective model fits ( black lines ) that include three components ( dotted lines ) : primary white dwarf , accretion disk with emission lines , and the secondary star.,height=624 ] the parameters of all three systems discussed here are similar to those of hs2331 + 3905 @xcite , sdss1339 + 4847 @xcite , and sdss1035 + 0551 @xcite , which were analyzed in a similar fashion , i.e. white dwarf temperatures in the range 1050013000k , donor stars later than m8 , disk temperatures around 6500k , and surface densities of a few @xmath26 .
follow - up observations of sdss1035 + 0551 revealed eclipses in the light curve , confirming the existence of a brown dwarf donor with @xmath27 @xcite .
the limits on the spectral types determined from the sdss spectra suggest that a substantial fraction of these objects could harbor brown dwarf secondary stars .
we obtained optical high - speed time - series photometry on suitable sdss cvs using the 3.5 m telescope at apache point observatory ( apo ) in new mexico , the 2.2 m telescope at calar alto observatory ( cao ) in spain , and the 3.5 m wiyn telescope at kitt peak national observatory in arizona .
we used the dual imaging spectrograph ( dis ) in imaging mode , mounted at the nasmyth focus of the 3.5 m telescope at apo .
the instrument design utilizes a dichroic that splits the beam of white light at 555 nm .
the blue and red portions of the beam are then incident on two different ccd cameras .
we present filterless light curves obtained using the blue ccd camera in this paper , sensitive in the wavelength range 350 - 555 nm .
the instrument has a read noise of about 4.6 electrons rms , a gain of 1.75 e / adu , and a plate scale of 0.4 arcsec / pixel at the 3.5 m telescope at apo .
we used windowing to read a small portion of the ccd in order to reduce the read time to 24s . including an additional overhead of 5s from ccd flushing ,
writing data to disk , and other software delays , we obtained a dead time of order 79s between exposures .
we used a standard iraf reduction to extract sky - subtracted light curves from the ccd frames using weighted circular aperture photometry @xcite .
at the 2.2 m cao telescope , we used the calar alto faint object spectrograph ( cafos ) to obtain filterless ccd photometry .
we used windowing and binning in order to reduce the read time and hence the dead time between exposures .
we applied a bias correction and flat - fielded the images within midas , and then used the source - extractor algorithm @xcite to perform aperture photometry on all objects in the field - of - view and determined corresponding light curves .
we obtained the photometric observations at the wiyn 3.5 m telescope using the orthogonal parallel transfer imaging camera ( optic ) , the prototype orthogonal transfer ccd imager @xcite .
it consists of two 2kx4k orthogonal transfer ccds mounted together in a single dewar .
the camera has a read noise of @xmath284 electrons rms , a gain of 1.45 e / adu , and a plate scale of 0.12 arcsec / pixel at the wiyn 3.5 m telescope .
we used optic in conventional mode ( see * ? ? ? * ) and all the time - series images were binned 2x2 , providing an 8s readout time between successive exposures .
we observed through a blue bandpass bg-39 filter ( @xmath1b+v ) for our observations .
we reduced all the data from the wiyn telescope in exactly the same manner as the apo data .
although the existing instruments at all the telescopes mentioned above allow short integration times , they are not entirely suitable to study pulsating white dwarfs because they include a substantial dead time between exposures .
an ideal instrument such as a blue - sensitive frame transfer ccd camera would not only allow short exposure times , but also include insignificant dead times between consecutive exposures , to enable data acquisition with high time resolution .
we indicate our journal of observations acquired at the three different telescopes in table[jour - obs ] .
ccccc apo 3.5 m & sdss0745 + 4538 & 14 oct 2005 11:29:10.458 & 0.83 & 15 + apo 3.5 m & sdss0745 + 4538 & 30 nov 2005 08:03:16.534 & 2.13 & 15 + apo 3.5 m & sdss0919 + 0857 & 01 dec 2005 10:47:54.323 & 2.17 & 15 + apo 3.5 m & sdss0919 + 0857 & 05 dec 2005 10:40:26.316 & 2.17 & 15 + cao 2.2 m & sdss0745 + 4538 & 20 jan 2006 23:18:54.403 & 3.83 & 25 + cao 2.2 m & sdss0745 + 4538 & 21 jan 2006 21:12:25.027 & 3.35 & 25 + cao 2.2 m & sdss0745 + 4538 & 22 jan 2006 20:03:04.435 & 0.86 & 35 + cao 2.2 m & sdss0745 + 4538 & 23 jan 2006 21:00:57.629 & 3.56 & 30 + cao 2.2 m & sdss0919 + 0857 & 24 jan 2006 03:26:17.578 & 2.95 & 30 + apo 3.5 m & sdss0745 + 4538 & 30 jan 2006
03:21:42.044 & 2.14 & 15 + apo 3.5 m & sdss0919 + 0857 & 20 feb 2006 03:16:44.490 & 3.80 & 15 + wiyn 3.5 m & sdss1711 + 3013 & 29 may 2006 08:44:06.79 & 2.50 & 25 + wiyn 3.5 m & sdss1711 + 3013 & 30 may 2006 07:12:25.10 & 3.21 & 25 + apo 3.5 m & sdss0919 + 0857 & 16 mar 2007 06:05:39.243 & 0.85 & 15 + apo 3.5 m & sdss0919 + 0857 & 18 mar 2007 02:08:02.664 & 3.81 & 10 + after extracting the light curves , we divided the light curve of the target star with a sum of one or more comparison stars ; we used brighter stars for the division whenever available as their light curves have lower noise .
after this preliminary reduction , we brought the data to the same fractional amplitude scale ( @xmath29 ) and converted the mid - exposure times of the ccd images to barycentric coordinated time ( tcb ; * ? ? ?
we then computed a discrete fourier transform ( dft ) for all the light curves up to the nyquist frequency .
we estimate an effective temperature of 11000k for the accreting primary white dwarf in sdss0745 + 4538 ( @xmath30=19.05 ) using its optical sdss spectrum and assuming @xmath12 g=8 ( see table[t - fitparameter ] ) .
we show the optical light curves and corresponding dfts of sdss0745 + 4538 data obtained at the 3.5 m telescope at apo and the 2.2 m telescope at cao in figure[lcft - cvza ] .
we indicate the periodicities observed on individual nights in table[per - cvza ] .
the longest g - mode pulsation period observed in cool non - interacting pulsating white dwarfs is @xmath11400s .
hence we expect that the photometric variations in the light curves with periods longer than 40min are indicative of the orbital period .
the 43.2min period observed on the 30th of january 2006 is merely a harmonic of the 86.5min period observed on the 20th of january 2006 . using a weighted average of both these periodicities , where the weights are inversely proportional to the 1@xmath31 uncertainties , we determine the longest observed period to be @xmath32min . as the orbital periods of most of the accreting pulsators fall in the range of 80 - 90min , indicative of their low rates of mass transfer ( see section 2 ) ,
we conclude that our measurement of @xmath32min is the orbital period of sdss0745 + 4538 .
light curves and discrete fourier transforms ( dfts ) of sdss0745 + 4538 data obtained using the 3.5 m telescope at apache point observatory and the 2.2 m telescope at calar - alto observatory .
we indicate the single pulsation period @xmath33 observed in each run , along with the first ( @xmath33/2 ) and second harmonics ( @xmath33/3 ) , indicative of its nonlinear pulse shape.,title="fig:",width=307 ] light curves and discrete fourier transforms ( dfts ) of sdss0745 + 4538 data obtained using the 3.5 m telescope at apache point observatory and the 2.2 m telescope at calar - alto observatory .
we indicate the single pulsation period @xmath33 observed in each run , along with the first ( @xmath33/2 ) and second harmonics ( @xmath33/3 ) , indicative of its nonlinear pulse shape.,title="fig:",width=288 ] lllll 14oct2005 & & * 1166@xmath2223 ; [email protected] * & [email protected] ; [email protected] & + 30nov2005 & & * 1246@xmath2211 ; [email protected] * & [email protected] ; [email protected] & + 20jan2006 & [email protected] ; [email protected] & * [email protected] ; [email protected] * & [email protected] ; [email protected] & [email protected] ; [email protected] + 21jan2006 & & * [email protected] ; [email protected] * & [email protected];[email protected] & [email protected] ; [email protected] + 22jan2006 & & * 1290@xmath2240 ; [email protected] * & & + 23jan2006 & & * [email protected] ; [email protected]*&[email protected] ; [email protected] & [email protected] ; [email protected] + 30jan2006 & [email protected] ; [email protected] & * [email protected] ; [email protected] * & [email protected] ; [email protected] & + we find that all the light curves show a period ( @xmath33 ) in the narrow range of 11661290s with a significantly high amplitude .
they also show the first harmonic ( @xmath33/2 ) of this dominant period @xmath33 , except for the short run on the 22nd of january 2006 .
we can even see the second harmonic ( @xmath33/3 ) of this period @xmath33 in our data on the 20th , 21st , and 23rd of january 2006 .
we will ignore these harmonics for now as they merely reflect the nonlinear pulse shape of the fundamental mode @xmath33 , and they are not linearly independent modes .
each light curve of sdss0745 + 4538 therefore shows evidence of a single independent period @xmath33 besides the orbital period . the value of this dominant independent period @xmath33 changes from 1166s to 1290s , as shown in table[per - cvza ] . the suggestive lack of stability in the observed period @xmath33 rules out the possibility that it represents the spin period of the white dwarf .
the spin period of a white dwarf in a cv has been shown to be extremely stable , and we expect the rate of change of period with time to be @xmath34 @xcite .
we now consider the possibility that the dominant period @xmath33 in each of the light curves of sdss0745 + 4538 represents a pulsation mode excited in the primary white dwarf .
we find that the observed range of periods 11661290s are consistent with nonradial g - mode pulsations in white dwarf stars .
furthermore we observe high amplitudes in the long period ( 6001200s ) zz ceti stars @xcite , which is also consistent with the observed amplitudes in sdss0745 + 4538 .
the pulse shapes of the long period zz ceti stars are mostly non - sinusoidal and therefore result in harmonics in the dfts , which is similar to the behavior shown by sdss0745 + 4538 .
we also see evidence of large amplitude modulation in the pulsation spectra of long period zz ceti stars even on timescales as short as a few days ( e.g. * ? ? ?
* ) ; we explain the phenomenon of amplitude modulation below .
the eigenfrequencies of a pulsating white dwarf , representative of its fundamental parameters and stellar structure , change only on evolutionary timescales @xmath35hz / s ( e.g. * ? ? ?
we can expect that accreting pulsating white dwarfs evolve faster than their isolated counterparts , but we still expect changes in the eigenfrequencies to occur on extremely slow timescales @xmath36hz / s ( see * ? ? ?
variable white dwarfs exhibit only a handful of all possible eigenfrequencies at any given time .
if new eigenmodes get excited in the star and previously excited eigenmodes are entirely suppressed , then the observed pulsation spectrum will show different frequencies .
this does not imply that the eigenfrequencies themselves are changing , but that the amplitude of excitation can vary on short timescales from zero to a finite value .
this amplitude modulation explains why we observe a different dominant period @xmath33 in each of the light curves of sdss0745 + 4538 .
the behavior of the optical variations in sdss0745 + 4538 is completely consistent with nonradial white dwarf pulsations .
we estimate an effective temperature of 13000k for sdss0919 + 0857 ( @xmath30=18.2 ) , uncertain by a few 1000k ( see table[t - fitparameter ] ) .
we show the optical light curves and corresponding dfts of sdss0919 + 0857 acquired using the 3.5 m telescope at apo and the 2.2 m telescope at cao in figure[lcft - cvzb ] .
we indicate the periods and amplitudes of the optical variations observed on individual nights in table[per - cvzb ] .
light curves and dfts of sdss0919 + 0857 data acquired using the 3.5 m telescope at apo and the 2.2 m telescope at cao .
we indicate the location of the period @xmath1260s and its first harmonic ( p/2 ) against the dfts computed for all the runs .
pulsations are clearly evident in the light curve from 18th of march 2007 shown separately.,title="fig:",width=288 ] light curves and dfts of sdss0919 + 0857 data acquired using the 3.5 m telescope at apo and the 2.2 m telescope at cao .
we indicate the location of the period @xmath1260s and its first harmonic ( p/2 ) against the dfts computed for all the runs .
pulsations are clearly evident in the light curve from 18th of march 2007 shown separately.,title="fig:",width=288 ] light curves and dfts of sdss0919 + 0857 data acquired using the 3.5 m telescope at apo and the 2.2 m telescope at cao .
we indicate the location of the period @xmath1260s and its first harmonic ( p/2 ) against the dfts computed for all the runs .
pulsations are clearly evident in the light curve from 18th of march 2007 shown separately.,height=96 ] @xcite determine the orbital period of sdss0919 + 0857 to be 84@xmath37min from radial velocity measurements .
the light curve variations longer than 30min are all suggestive of being either direct measurements or harmonics of this period .
most of these measurements are consistent with each other , considering that the uncertainties are under - estimated .
we compute a weighted average for the orbital period to be 90.9@xmath38min , where the weights are inversely proportional to the 1@xmath31 uncertainties .
we boost the error bar to 7.3min to make the average value consistent with all the different measurements .
this step is necessary to derive a realistic estimate of the uncertainty in measuring the orbital period ; the observing runs are too short to determine the orbital period reliably and hence the least squares uncertainties are severely under - estimated .
our determination of the orbital period then becomes 91@xmath39min , which is about seven minutes longer than the determination of @xcite , but consistent within the uncertainties .
lllllll 01dec2005 & & [email protected] ; [email protected] & [email protected] ; [email protected] & & & + 05dec2005 & & & & & & + 24jan2006 & & [email protected] ; [email protected] & & & & [email protected] ; [email protected] + 20feb2006 & [email protected] ; [email protected] & [email protected]
; [email protected] & & & & + 16mar2007 & & & & & & + 18mar2007 & [email protected] ; [email protected] & [email protected] ; [email protected] & & & [email protected] ; [email protected] & + all , but one , of the light curves indicate a short period @xmath33 near 260s , as shown in table[per - cvzb ] .
measurements from the 5th of december 2005 and the 24th of january 2006 indicate a value near 259.9s , while the other measurements indicate a distinct value near 261.3s .
this suggests that the 260s period is a doublet ; sampling different phases of the beat cycle where one mode may dominate over the other can explain the slightly different period measurements . in the short period zz ceti stars ( 200300s ) close to the blue edge of the non - interacting zz ceti instability strip
, we often observe multiplet structure in the excited modes due to rotation or magnetic field .
these stars typically show lower amplitudes in the range of 120mma @xcite , which is consistent with the amplitudes observed for the 260s period .
our present constraint on the splitting of the 260s mode is about 20.6@xmath40hz .
if the splitting were caused by rotation , this would imply a spin period of @xmath11.8days , which is too slow for an accreting white dwarf .
it is therefore much more likely that the splitting is caused by a magnetic field @xmath41 g ( see * ? ? ?
we can justify that the mode was not observed on the 1st of december 2005 due to beating of the closely spaced frequencies .
the amplitude modulation from 6.8mma on the 5th of december 2005 to 15.8mma on the 16th of march 2007 lends support to this idea .
although we find no evidence to contradict our idea that the 260s period is a nonradial g - mode , we are unable to eliminate the remote possibility that it could be the spin period of the white dwarf . we expect to measure the spin period of a rapidly rotating white dwarf in a cv whenever there are features on the surface of the white dwarf , such as a hot spot .
such a hot spot typically arises in accreting white dwarfs with a strong magnetic field , where the accretion flow close to the white dwarf becomes field - aligned .
the accretion stream then funnels to one or both magnetic poles .
the stream undergoes a shock front and radiates x - rays which heat the white dwarf surface near the pole(s ) .
there are two kinds of cvs that show hot spots : intermediate polars with magnetic fields of 110 mg , usually with some outer accretion disk , and polars with fields @xmath715 mg devoid of a disk .
since the optical spectrum of sdss0919 + 0857shows double - peaked lines , typical of an accretion disk , it is not likely to be a polar cv .
the spectrum does not show any signs of zeeman splitting , suggesting that the field is smaller than 1 mg ( see * ? ? ? * ) .
also , there is no x - ray source associated with sdss0919 + 0857 ( see * ? ? ? * and references therein ) .
these constraints reduce the possibility that sdss0919 + 0857 is an intermediate polar , and the 260s period represents the spin period of the white dwarf caused by a rotating hot spot .
however we are not completely certain because @xcite claim to observe the spin period of 1.12min in the recently discovered accreting pulsating white dwarf hs2331 + 3905 ; its optical spectrum does not show any zeeman splitting and it is not associated with any x - ray source either .
we had initially dismissed the 144s period observed on the 24th of january 2006 as noise due to flickering .
but we find that this period is perhaps revealing the interaction of the orbital period with the 260s period .
let @xmath42 be the frequency associated with the harmonic of the orbital period 41.8min observed on the 24th of january 2006 .
let @xmath43 and @xmath44 be the frequencies associated with the 259.93s and 144.09s periods respectively .
we find that within uncertainties @xmath45 .
this implies that @xmath44 could be a linear combination frequency , however its amplitude is only 2.5 times the average noise amplitude , making it less reliable than a 3@xmath31 detection .
we can not explain why we do not observe @xmath46 , and why the 144s period is only visible during one set of observations , if it is indeed real .
this does not help us resolve the dilemma of whether the 260s period represents the spin period or a pulsation mode , as both these phenomena could potentially interact with the orbital period through tides .
our observations on the 1st of december 2005 do not show the 260s period inspite of a suitably low noise level ( see figure[lcft - cvzb ] ) .
it is possible that the accretion is clumpy at these low rates of mass transfer . if the accretion stopped for some time , the hot spot could cool off , explaining the absence of the 260s period during our first run under this scenario .
we can explain the observed 260s period with two different models : nonradial pulsation and rotation .
we adopt the model of nonradial pulsation here because it is more likely .
we report a non - variable sdss1711 + 3013 ( @xmath30=20.25 ) , which we observed twice during our search , obtaining useful non - variability limits related to the absence of pulsations in the primary white dwarf .
note that even cvs that show variability from flickering , spin or orbital period modulations , would still count as non - variables in this context .
we show the light curves and dfts of sdss1711 + 3013 acquired using the wiyn 3.5 m telescope in figure[lcft - nova ] .
we indicate the periods and amplitudes of the optical variations related to the orbital period in table[per - nova ] .
we show the light curves and corresponding dfts of sdss1711 + 3013 data obtained using the 3.5 m wiyn telescope .
we also indicate the pulsation related non - variability limit in mma , subsequent to the nov designation.,title="fig:",width=288 ] we show the light curves and corresponding dfts of sdss1711 + 3013 data obtained using the 3.5 m wiyn telescope .
we also indicate the pulsation related non - variability limit in mma , subsequent to the nov designation.,title="fig:",width=288 ] llll 29 may 2006 & & [email protected] ; [email protected] & @xmath47mma for periods within 831430s ( nov7 ) + 30 may 2006 & [email protected] ; [email protected] & [email protected] ; [email protected] & @xmath45mma for periods within 771470s ( nov5 ) + the dft of sdss1711 + 3013 acquired on the 29th of may 2006 exhibits substantial power at low frequencies , and we fit the highest unresolved peak to determine the harmonic of the orbital period @xmath839.4min .
the light curve from the next night distinctly shows both the orbital period and its harmonic at 83.3min & 38.8min respectively .
our weighted average of these three measurements is [email protected] , using weights that are inversely proportional to the 1@xmath31 uncertainties .
we artificially boost the error bar to 4.5min for consistency with all the measurements .
hence our determination of the orbital period becomes 79@xmath225min .
note that both light curves from figure[lcft - nova ] show a distinctive feature : the orbital modulation consists of a @xmath13035min lower amplitude feature followed by a @xmath150min relatively higher amplitude feature .
sdss1711 + 3013 was not observed to vary ( nov ) , and we designate it as nov@xmath47 , where @xmath47 represents the non - variability limit determined from the dft , similar to the scheme for non - variables within and around the zz ceti instability strip ( see * ? ? ?
the highest peak in the dft essentially defines the detection threshold or the non - variability limit , provided it is not related to the orbital period or also present in the dfts of the reference stars . in that case , we apply the same test to the second - highest peak , and so on , until we can determine the highest peak , truly representative of the non - variability limit .
the data acquired using the wiyn 3.5 m telescope on the 29th of may 2006 ( top right panel , figure[lcft - nova ] ) allow us to constrain the pulsation amplitude below 7mma for periods in the range of 831430s .
the data from the 30th of may 2006 ( bottom right panel , figure[lcft - nova ] ) restrict the pulsation amplitude to below 5mma for periods in the range of 771470s .
we conclude that there are no pulsation related periodicities in the range of about 851400s that are higher than the limiting amplitude of 0.5% ; we designate this non - variable as nov5 .
we estimate an effective temperature of 10500k for the accreting primary white dwarf in sdss1711 + 3013 using its optical sdss spectrum ( see table[t - fitparameter ] ) . since most of the known accreting white dwarf pulsators cluster in the temperature range of 1450015000k @xcite ,
we wonder if the non - variable sdss1711 + 3013 is too cool to lie within the instability strip(s ) for variable accretors . @xcite
determine an effective temperature of 10500k for the accreting white dwarf pulsator hs2331 + 3905 .
however given the uncertainty of @xmath12000k in our temperature estimate , sdss1711 + 3013 could easily be substantially cooler than hs2331 + 3905 .
the light curves of sdss0745 + 4538 , sdss0919 + 0857 , and sdss1711 + 3013 show optical variations indicative of the orbital period of these binary systems .
we determine an orbital period of @xmath32min for sdss0745 + 4538 , @xmath48min for sdss0919 + 0857 , and @xmath49min for sdss1711 + 3013 using our data .
we also observe a single independent period in the light curves of sdss0745 + 4538 in the range of 11661290s , which varies from run to run .
the spin of the white dwarf is expected to be highly stable and can not explain the observed variations .
we deduce that the variations in the light curves of sdss0745 + 4538 with long periods , high amplitudes of 4.57% , non - sinusoidal pulse shapes , and substantial amplitude modulation are completely consistent with nonradial g - modes , similar to those observed for the non - interacting long period cool zz ceti stars .
we observe a persistent period close to 260s in five out of six of the light curves of sdss0919 + 0857 .
we adopt the explanation that this period is a nonradial pulsation mode with doublet structure and consequent beating , similar to the modes excited in the short period zz ceti stars .
the observed amplitudes are also consistent with the small amplitudes observed in the short period zz ceti stars , found close to the blue edge of the zz ceti instability strip .
but we can not rule out the remote possibility that the 260s period could be the spin period of the white dwarf without additional data that span a long timebase . even after observing sdss1711 + 3013 twice
, we did not find any pulsation related periodicities in the range of 851400s that are higher than the limiting amplitude of 0.5% ( nov5 ) .
we therefore conclude that sdss1711 + 3013 does not exhibit nonradial pulsations .
our temperature estimate of 10500k for this accreting white dwarf suggests that it may perhaps be too cool to pulsate .
this discovery paper brings the total number of known accreting white dwarf pulsators to eleven .
support for this work was provided by nasa through the hubble fellowship grant hst - hf-01175.01-a awarded by the space telescope science institute , which is operated by the association of universities for research in astronomy , inc .
, for nasa , under contract nas 5 - 26555 . we thank nsf for the grant ast-0607840 .
btg was supported by a pparc advanced fellowship .
aa thanks the royal thai government for a studentship . based on observations obtained with the apache point observatory 3.5-meter telescope , which is owned and operated by the astrophysical research consortium and on observations collected at the centro astronmico hispano alemn ( caha ) at calar alto , operated jointly by the max - planck institut fr astronomie and the instituto de astrofsica de andaluca ( csic ) . also based on observations obtained at the wiyn observatory , which is a joint facility of the university of wisconsin - madison , indiana university , yale university , and the national optical astronomy observatories
we thank the referee for helping us make this manuscript a better paper . | we report the discovery of two new accreting pulsating white dwarf stars amongst the cataclysmic variables of the sloan digital sky survey : sdssj074531.91 + 453829.5 and sdssj091945.10 + 085710.0 .
we observe high amplitude non - sinusoidal variations of 4.57% at a period close to 1230s in the optical light curves of sdssj074531.91 + 453829.5 and a low amplitude variation of 0.71.6% near 260s in the light curves of sdssj091945.10 + 085710.0 .
we infer that these optical variations are a consequence of nonradial g - mode pulsations in the accreting primary white dwarfs of these cataclysmic variables .
however we can not rule out the remote possibility that the 260s period could be the spin period of the accreting white dwarf sdssj091945.10 + 085710.0 .
we also uncovered a non - variable sdssj171145.08 + 301320.0 during our search ; our two observing runs exclude any pulsation related periodicities in the range of 851400s with an amplitude @xmath00.5% .
this discovery paper brings the total number of known accreting white dwarf pulsators to eleven . |
cryogenic radiation - detectors that utilize ionization , phonon and / or scintillation measurements are being used in a number of experiments .
both the cryogenic dark matter search ( cdms ) @xcite and edelweiss @xcite dark matter search utilize silicon and / or germanium targets to detect recoils of radiation in the target masses .
a combination of ionization and phonon readout is used to provide discrimination of gamma- and neutron - recoil types .
the cresst dark matter search utilizes cawo@xmath0 targets and readout scintillation and phonon signal to discriminate between recoil types .
the advantage of reading out both phonon and ionization ( or scintillation ) signals comes about from the differing ratios of ionization and phonon energy or scintillation and phonon energy created in electron- and nuclear - recoils in the detectors .
the ratio of these two energies leads to a powerful discriminator for the experiment s desired recoil type .
both the ionization and phonon readout can be used to generate position estimators for the initial radiation interaction , leading to fiducial volume selection . in the ionization signal
this is generally accomplished by instrumenting different parts of the detector with independent readout channels and vetoing events with large signal contribution outside of the desired fiducial volume . in the phonon signal it
is generally required to measure the early , athermal component of the phonon signal which still retains a position dependent component .
the physics required to accurately model these detectors is presented in this paper along with appropriate numerical tricks that are useful for an efficient detector monte carlo .
this paper proceeds with a review of radiation interactions , charge transport physics , phonon transport physics , instrumentation .
monte carlo techniques and relevant physical constants are included where appropriate .
this paper will focus on the use of silicon and germanium detector masses , both of which are group iv semiconductors .
however there are other relevant materials in use such as calcium tungstate ( cawo@xmath0 ) which leads to a small loss of generality . the cryogenic dark matter search @xcite utilizes silicon and germanium detectors to search for weakly interacting massive particle ( wimp ) dark matter @xcite candidates .
the silicon or germanium nuclei provide a target mass for wimp - nucleon interactions .
simultaneous measurement of both phonon energy and ionization energy provide a powerful discriminator between electron - recoil interactions and nuclear - recoil interactions .
background radiation primarily interacts through electron - recoils whereas a wimp signal would interact through nuclear - recoils .
the experiment is located in the soudan mine , mn , u.s.a . the most recent phase of the cdms experiment has involved fabrication , testing and commissioning of large , 3 inch diameter , 1 inch thick [ 100 ] germanium crystals .
the cdms - izip ( interleaved z dependent ionization and phonon ) detectors are 3 inches in diameter and 1 inch thick with a total mass of about 607 grams @xcite .
the izip detector utilizes both anode and cathode lines on the same side of the detector similar to a micro - strip gas chamber ( msgc ) @xcite as shown in figure [ fig : izip ] and [ fig : iziptes ] . unlike an msgc
however , there is a set of anode and cathode lines on both sides of the detector .
this ionization channel design is used to veto events interacting near the detector surfaces .
an amorphous silicon layer , deposited under the metal layers , increases the breakdown voltage of the detectors .
the total izip aluminum coverage is @xmath14.8% active and @xmath11.5% passive per side .
c ( right ) close - up view of the izip phonon channel and ionization channel ( thin lines in between the phonon sensors ) .
the phonon channel is held at ground and the ionization channel is held at @xmath22 v for the top ( bottom ) surfaces.,width=264 ] ( right ) close - up view of the izip phonon channel and ionization channel ( thin lines in between the phonon sensors ) .
the phonon channel is held at ground and the ionization channel is held at @xmath22 v for the top ( bottom ) surfaces.,width=264 ]
when using a monte carlo of a detector , it is often helpful or necessary to have a numerical model of radiation interactions in the detector
. many readers will find it valuable to use separate modeling software such as geant4 @xcite .
a brief description of these interactions follows .
low energy gamma - rays ( x - rays ) predominantly interact via photoelectric absorption in which all of the gamma - ray energy is deposited in a single interaction location .
high energy gamma - rays interact via compton scattering in which some of the gamma - ray s initial energy is transferred to an electron and the gamma - ray continues along a different trajectory with reduced energy
. the gamma - ray will generally continue to scatter until it leaves the detector volume or terminates with a photoelectric absorption . in silicon ( germanium ) , for photon energies greater than 60 ( 160 ) kev
, compton scattering dominates @xcite .
both of these electron interactions result in a high energy electron being produced which then undergoes a rapid cascade process resulting in a large number of electron - hole pairs @xcite .
this initial cascade process ceases around the scale of the mean electron - hole pair creation energy ( @xmath3 ) resulting in an expected number of electron - hole pair @xmath4 . due to correlations in the cascade process ,
the variance in the number of electron - hole pairs is reduced , relative to poisson statistics , and given by @xmath5 , where @xmath6 is the fano factor @xcite .
these high energy electron - hole pairs will then shed phonons until they reach the semiconductor gap energy @xmath7 which results in the fraction @xmath8 of energy in prompt phonons and the remainder in the electron - hole pair system .
neutrons that interact in the detector bulk will knock an ion out of its lattice site and displace it to some other location in the crystal .
this high energy ion will interact with both the lattice ions and / or valence electrons , with competing cross sections , before reaching some other location in the crystal .
interactions with the lattice ions can be described to first approximation as rutherford scattering @xcite with differential energy loss per unit length @xmath9 , where @xmath10 is the ion s velocity @xcite . for interactions with valence electrons
, the number of electron states which can be excited to an accessible state outside of the fermi sphere scales like velocity @xmath10 hence the differential energy loss per unit length scales like @xmath11 @xcite .
a description of this process is shown in figure [ fig : elecionscatter ] .
there are important screening and velocity dependent cross sections in both energy loss lengths @xcite but to first approximation these equations show the velocity dependent competition in the two scattering rates .
whichever interaction occurs first , there is generally a cascade of many scattering events resulting in a larger amount of energy being deposited in the ion lattice compared to gamma - ray interactions . for historical reasons
, the reduced amount of energy in the electron - hole ( ionization ) system , compared to an equal energy deposition by a gamma - ray , has brought about the term _ nuclear quenching _ to describe the reduced ionization signal .
the details of the cascade and reduction in the number of electron - hole pairs is described by lindhard theory @xcite and given as @xmath12 where @xmath13 , @xmath14 and @xmath15 @xcite .
the recoil energy is @xmath16 and given in units of kev , @xmath17 is the atomic number and @xmath18 is the atomic mass .
beta radiation represents another class of electron - recoil interactions .
the attenuation lengths are much shorter , however , resulting in a class of events sometimes referred to as _ surface interactions_. these surface interactions can result in signals that differ from bulk events of the same energy and electron- / nuclear - recoil type .
for example , the events may be located in regions of large electric fringing fields , directing charges away from readout electrodes or near mounting hardware that absorbs scintillation light .
description of ion - electron potential interaction process which shows that for ion velocity much less than the fermi velocity ( @xmath19 ) the number of electron states that an ion can interact with scales like @xmath10.,width=529 ]
phonons , in the context of this review , are quantized vibration modes that exist in periodic structures such as silicon and germanium crystals .
they are excitations which , in the lattice , and in materials sufficiently cooled that charge carriers are frozen out , mediate thermal transport .
they are described by the hamiltonian @xmath20 where @xmath21 is the mass of the lattice atoms and @xmath22 is the frequency of oscillations about the center of the harmonic potential between an atom and its nearest neighbor atom @xcite . in one dimension , the solution to the time - independent schrodinger equation @xmath23 yields the different oscillation modes at atom @xmath24 @xmath25 where @xmath26 here , the wave number @xmath27 , @xmath28 is the lattice spacing and @xmath29 are the number of lattice sites . in monte carlo ,
phonons are simply treated as non interacting particles with decay and mass defect scattering properties described in the remainder of this chapter . in general
they have a nonlinear dispersion relationship ; however , due to rapid down conversion to lower frequencies they spend most of their time in an energy region with linear dispersion relationship . hence
, a linear dispersion relation is sufficient for phonon transport modeling in these detectors . immediately after the recoil , a population of prompt phonons exist in the immediate vicinity of the interaction point .
particular details of the frequency distribution and mode population are not well known but we can make a few deductions , explained in more detail in the following sections , that will lead us an initial distribution for monte carlo .
anharmonic decay , due to nonlinear terms in the elastic coupling between adjacent lattice ions , causes the phonons to rapidly down convert into a lower frequency distribution .
this process allows us to start a monte carlo with any high frequency distribution and details of the distribution will rapidly be lost ; we use the debye frequency as a naive starting point . isotope scattering , which also occurs at a high rate for a high frequency phonons , causes the phonons to obtain their equilibrium mode density ; we use the equilibrium mode density as a naive starting point .
the approximations are good in the sense that the detector s phonon response is insensitive to variations in the distributions .
they are valid since the detectors are large compared to the initial characteristic interaction lengths of phonons .
furthermore , later generations of phonons are ballistic and timing information in the measured phonon pulses is determined by the detector geometry and the loss rate of phonons at surfaces .
the so called phase velocity surfaces represent the direction dependent phonon phase velocity @xmath30 .
in general they are given by the eigenvalue equation [ eq : phasevelocity ] .
@xmath31 where @xmath32 is the crystal s mass density , + @xmath22 is the phonon frequency , + @xmath33 is a component of the polarization vector @xmath34 , + @xmath35 are components of the elastic constant tensor and , + @xmath36 is a component of the phase velocity vector @xmath37 @xcite .
not all of the elastic constants are independent , reducing via a voigt contraction @xcite the number that we need to keep track off .
additionally , symmetries in a cubic crystal allow for further reduction in components and we can define three independent constants @xmath38 , @xmath39 and @xmath40 .
this contraction simplifies equation [ eq : phasevelocity ] significantly @xcite and we are left solving matrix [ eq : phasevelocitymatrix ] for its eigenvectors and eigenvalues .
@xmath41 the eigenvectors represent the three polarization vector directions and the three eigenvalues equal @xmath42 for the longitudinal , slow - transverse and fast - transverse modes .
the anisotropy in the cubic silicon crystal leads to the phase velocity surfaces being non - spherical ( see figure [ fig : sidispersion ] ) .
the phase velocities are used to determine both the group velocities ( section [ sect : groupvelocity ] ) and also the isotope scattering rates .
phase velocities in silicon .
the distance from the origin represents the phase velocity speed ( in m / s ) .
the first three plots are views from the north pole .
the fourth view containing all three surfaces is offset from the north pole . in the plots all modes
are equally populated , which does not reflect the actual mode populations.,width=529 ] phonon group velocities are found by solving @xmath43 the slight lack of sphericity in the phase velocity surfaces ( see figure [ fig : sidispersion ] ) has a very dramatic effect on the transverse phonon group velocities @xcite ( see figure [ fig : sigroupvelocity ] ) .
the longitudinal phonon s group velocity is only mildly affected .
energy is focused in the direction of heavy banding and leads to the term _
phonon focusing_. the point density in the plots is misleading as the three modes are shown to be equally populated .
isotope scattering including anisotropic scattering rates ( see section [ sect : isotopescatter ] ) leads to the phonon modes in silicon being populated as follows : slow - transverse ( 55% ) , fast - transverse ( 35% ) and longitudinal ( 10% ) .
group velocities in silicon .
energy is focused in the direction of heavy banding and leads to the term _
phonon focusing_. the distance from the origin represents the speed that phonon energy is carried through the crystal ( in m / s ) . in the plots
all modes are equally populated , which does not reflect the actual mode populations.,width=529 ] phonons scatter off mass defects in the crystal ( see figure [ fig : isotopescatter ] ) .
additionally , they can change modes .
the bulk scattering rate is given by @xmath44,\ ] ] where @xmath45 is the phonon frequency and @xmath46 is a scattering rate constant @xcite ( see table [ tab : phononsimconst ] ) .
the scattering rate for individual phonons is given by @xmath47 where @xmath48 is the final state phonon frequency in hz , @xmath49 is the polarization vector , @xmath50 represents the initial phonon and @xmath51 represents the outgoing phonon @xcite .
it is the dot product in equation [ eq : microisotopescatter ] which allows mode mixing and the denominator which ensures the correct populations in the ratios . in silicon the populations when including anisotropic scattering rates are slow - transverse ( 55% ) , fast - transverse ( 35% ) and longitudinal ( 10% ) .
the standard treatment is to determine if a phonon isotope scatters via @xmath52 and then determine its polarization and direction via @xmath53 .
after the initial anharmonic decay has settled down , isotope scattering dominates .
this process is unfortunately computationally expensive due to sample - rejection techniques @xcite and after several iterations an isotropic scattering process can be used for individual phonons with little loss of modeling accuracy
. phonons isotope scatter off mass defects in the crystal .
equation [ eq : microisotopescatter ] gives the individual phonon scatter rates .
@xmath54 is the group velocity and @xmath49 is the polarization vector.,width=377 ] nonlinear terms in the elastic coupling constants cause a longitudinal phonon to down convert to two lower energy phonons ( see figure [ fig : anharmonicdecay ] ) .
the bulk decay rate is given by @xmath55,\ ] ] where @xmath45 is the phonon frequency and @xmath18 is a decay rate constant @xcite ( see table [ tab : phononsimconst ] ) .
the decay rate for transverse phonons is negligible @xcite .
the three body problem requires that both energy and momentum are conserved .
these conditions make an exact solution computationally prohibitive for large amounts of phonons . longitudinal phonons decay due to nonlinear terms in the elastic coupling constants .
@xmath54 is the group velocity and @xmath49 is the polarization vector.,width=377 ] to allow computations to proceed in a finite time , an exact solution to anharmonic decay is abandoned and an isotropic approximation is used ( the full anisotropic phase velocities and group velocities are still easily used for isotope scattering and phonon transport ) .
the energy distribution calculations are still difficult but fortunately have already been carried out @xcite . once the energies have been determined , calculating the resultant scattering angles based on energy and momentum conservation is fairly straightforward . due to the different energy - momentum dispersion relations for the longitudinal and transverse phonons ,
there are two different decay branches , @xmath56 and @xmath57 .
the @xmath56 energy distribution is given by @xmath58 \left[1 + x^2 - \delta^2 ( 1-x)^2\right]^2,\ ] ] where @xmath59 , + @xmath60 and , + @xmath61 .
this approximation results in the outgoing phonons having an angular displacement from the initial phonon given by @xmath62 @xmath63 the @xmath57 energy distribution is given @xmath64 ^ 2,\ ] ] where @xmath65 , + @xmath60 , + @xmath66 , + @xmath67 $ ] , + @xmath68 , + @xmath69 and , + @xmath70 .
+ the constants @xmath53 and @xmath71 are the lam constants and @xmath72 and @xmath53 are third order elastic constants in an isotropic model ( there is additionally a third independent elastic constant @xmath73 but it drops out of the equations ) .
this approximation results in the outgoing phonons having an angular displacement from the initial phonon given by equations [ eq : l2ttangle1 ] and [ eq : l2ttangle2 ] .
@xmath74 @xmath75 where the trivial substitution @xmath76 is made for the second phonon .
with these closed - form energy densities and scattering angles , plots can be generated to aid understanding of these events ( see figures [ fig : anhenergyge ] and [ fig : anhanglege ] ) .
these angles are relative the initial momentum vector @xmath77 and need to be converted into the monte carlo coordinate system .
polar coordinates are useful and angles are provided in this system .
in addition to rotation angles @xmath78 and @xmath79 ( or @xmath80 and @xmath81 ) an additional azimuth angle , relative to the initial momentum vector @xmath77 , and in the isotropic approximation is randomly distributed from @xmath82 $ ] , is specified as @xmath83 . in terms of initial elevation and azimuth angles @xmath84 and @xmath85 , scattering angles @xmath80 and @xmath81 and azimuth scattering angle @xmath83 , the final angles @xmath86 and @xmath87 that describe the phonon momentum vectors @xmath88 are @xmath89\ ] ] and @xmath90.\ ] ] the final angles @xmath91 for the other phonon momentum vector @xmath92 are found by replacing @xmath80 with @xmath81 and @xmath83 with @xmath93 . c resultant angles in longitudinal phonon decay in germanium .
the two branches @xmath94 and @xmath95 are shown ; the plotted distribution is indicated in bold face in the legend.,width=377 ] + + resultant angles in longitudinal phonon decay in germanium .
the two branches @xmath94 and @xmath95 are shown ; the plotted distribution is indicated in bold face in the legend.,width=377 ] eventually the phonons will interact at surfaces where they are instrumented , reflected back into the crystal , down converted to a lower energy or are lost to the environment .
phonons can reflect either specularly or diffusively from the surfaces .
specular reflection can occur on smooth , untreated semiconductor wafer surfaces .
it is the simplest to describe with incident and reflected angles relative to the normal equal , @xmath96 .
diffusive scattering is also common on surfaces that have been damaged or roughened during fabrication .
in the ideal case the scattering angle is described by lambert s cosine law where the angular distribution scales like @xmath97 where the angle @xmath98 is measured relative to the normal .
scattering surfaces that satisfy lambert s cosine law scatter phonons isotropically regardless of their incident angle .
generally diffusive scattering has been found to be a good model for phonon - surface reflections , likely due to some small roughness in the surface .
phonons are strictly - speaking eigenstates of a hamiltonian that describes an infinitely large , periodic lattice .
this description necessarily breaks down at the detector surfaces and there is some probability that phonons down convert to lower energy daughters at the surface .
the details of this process would be highly material / surface treatment dependent but the probability of this occurring for a particular phonon - surface interaction will be small for a high purity crystal .
it is generally easiest to tune this probability by running numerous monte carlo to find the best value . in the cdms detectors we have measured a loss of @xmath10.1% for each phonon - surface interaction @xcite .
it is the goal of the experimental setup to absorb the phonons into some sensor and provide instrumentation into a data acquisition system .
the details for the phonon - sensor interaction probability are again complicated and highly depend on the type of absorber and attachment / fabrication details
. acoustic mismatch theory provides a good starting point for analytic calculations and can be performed over both normal and non - normal incidence angles .
the relevance of these calculations can be lost however when detector to detector variations are considered .
additionally the angular dependence of such calculations can be washed out when integrating over a distribution of phonon incident angles and phonon energies . in practice
it is usually again easiest to tune this probability by running numerous monte carlo and identifying the best fit value . in the cdms detector
there is additionally an amorphous silicon dielectric in between the crystal and thin metal films ; we have empirically matched a phonon - asi - aluminum interaction ( details of the interaction are discussed in section [ sec : qpdc ] ) probability of @xmath133% with the remaining 67% diffusively scattering back into the crystal @xcite .
for any well designed detector , phonon absorption into the instrumentation sensors will dominate over other loss processes allowing the probability to be tuned by matching pulse decay times .
it would be grossly inefficient to run all phonons with the same time step .
therefore we generate scattering times according to the distributions in sections [ sect : isotopescatter ] and [ sect : anharmonicdecay ] . the scattering and decay probabilities go like @xmath99 where @xmath100 is a combination of isotope scattering and anharmonic decay rates as given by matthiessen s rule @xmath101 .
this scattering time has to be compared with the time that the phonon will take to interact with a surface and the event that occurs soonest will be chosen for each individual phonon .
if it is determined that the bulk interaction time is less than the surface interaction time , then it must be determined which event occurs based on their relative , frequency dependent rates .
this can be done by drawing a uniform random number @xmath102 and comparing the rate of the process in question to the total rate .
for example , if @xmath103 then an anharmonic decay is selected . only in rare circumstances
will a uniform random number @xmath102 be needed in monte carlo without some transformation .
often we are trying to draw the number @xmath104 out of the probability distribution function ( pdf ) @xmath105 .
an efficient method for transforming @xmath102 to the desired probability distribution function @xmath105 is to integrate @xmath105 to find the cumulative distribution function @xmath106 .
the cumulative distribution function ( cdf ) @xmath6 has the desirable property that it is bounded by @xmath107 $ ] as is @xmath102 .
the cdf @xmath6 is then inverted and solved at @xmath102 to determine @xmath105 , @xmath108 @xcite . as an example , we first considering the bulk interaction rate where the probability of having an interaction is @xmath109 . after integration and inversion , the randomly generated scattering time is given by @xmath110 as a second example we can consider diffuse scattering off the detector walls . in a spherical coordinate system where @xmath98 represents the angle from the surface normal , the pdf in this coordinate is @xmath111 .
this is also easily integrated and inverted to yield @xmath112 .
care must be taken in this example however since @xmath113 is defined over the domain [ -1,1 ] which modifies to @xmath114 .
there are times when a pdf can not be analytically integrated to yield a cdf or the cdf can not be inverted .
this is an unfortunate situation since an expensive rejection technique is required .
this technique involves drawing a pair of uniform random numbers @xmath115 $ ] where @xmath116 and @xmath117 . if @xmath118 then @xmath119 is retained , otherwise @xmath119 is rejected and the process is repeated .
the inefficiency of this method is related to the area coverage @xmath120 and @xmath119 will be successfully drawn in one of @xmath121 attempts .
the rejection method can be improved however for _ static _ distributions that do not change during the monte carlo run .
an example includes diffusive scattering off of the side walls of the detector when using a spherical coordinate system . in this case , the @xmath122 jacobian results in the pdf @xmath123 for which @xmath6 can not be found analytically .
the pdf can be integrated numerically to generate a cdf which is subsequently inverted .
the process lacks a certain degree of elegance but is significantly more efficient than using a rejection method .
table [ tab : phononsimconst ] lists numerous constants that are used in the phonon simulations .
they define the propagation dynamics , scattering and decay rates and energy carrier statistics .
.numerical constants for phonon simulations [ cols="<,^,^,^,^,^",options="header " , ]
this paper has covered many physics and monte carlo topics in cryogenic radiation - detectors that utilize phonon and ionization readout .
the bulk response is different for gamma and neutron interactions , the former producing a larger ratio or ionization energy .
propagation of both the phonons and charge carriers is anisotropic , the phonons transport dependent on dispersion relations and the charge carrier transport dependent on mass tensors .
phonon interactions are complicated at the surface by electron - phonon downconversion requiring additional monte carlo effort .
the tes readout simulation is relatively straightforward relying on basic thermal and electrical processes .
various numerical techniques including pdf sampling and tricks for improving efficiency have been discussed .
monte carlo of detectors is a valuable tool for detector design , characterization and data analysis
. often counter intuitive results can be quickly discovered in monte carlo which aids in prioritizing laboratory r&d efforts .
furthermore , monte carlo can help to explain features in data , or at least rule out models .
detailed monte carlo studies in cdms @xcite and edelweiss @xcite detectors can be found in the literature .
these studies include discussion of quantities which are relevant in the tuning and their effects on phonon pulse shape , charge transport , tes phase - separation and tes noise . with the power of monte carlo and
increased use of low temperature cryogenic detectors , a general package is becoming more appealing .
the geant4 collaboration is beginning to implement phonon and charge transport modules into the geant4 toolkit and will expand the benefits of this research to others outside the cryogenic detector community @xcite .
i would like to thank the entire cdms collaborators for valuable discussions over the years , especially adam anderson , paul brink , blas cabrera , enectali figueroa - feliciano , scott hertel , peter kim , kevin mccarthy , matt pyle , bernard sadoulet , kyle sundqvist and betty young .
bc generously provided his note on `` electron - phonon scattering '' , which was adapted for this review .
km has been an invaluable contributor to the cdms monte carlo studies cited in the references and pk was invaluable in tirelessly running monte carlo on the slac computing farm .
d. n. spergel , r. bean , o. dor , m. r. nolta , c. l. bennett , j. dunkley , g. hinshaw , n. jarosik , e. komatsu , l. page , h. v. peiris , l. verde , m. halpern , r. s. hill , a. kogut , m. limon , s. s. meyer , n. odegard , g. s. tucker , j. l. weiland , e. wollack , and e. l. wright _ the astrophysical journal supplement series _ * 170*(2 ) , p. 377
, 2007 .
m. tegmark , m. a. strauss , m. r. blanton , k. abazajian , s. dodelson , h. sandvik , x. wang , d. h. weinberg , i. zehavi , n. a. bahcall , f. hoyle , d. schlegel , r. scoccimarro , m. s. vogeley , a. berlind , t. budavari , a. connolly , d. j. eisenstein , d. finkbeiner , j. a. frieman , j. e. gunn , l. hui , b. jain , d. johnston , s. kent , h. lin , r. nakajima , r. c. nichol , j. p. ostriker , a. pope , r. scranton , u. c. v. seljak , r. k. sheth , a. stebbins , a. s. szalay , i. szapudi , y. xu , j. annis , j. brinkmann , s. burles , f. j. castander , i. csabai , j. loveday , m. doi , m. fukugita , b. gillespie , g. hennessy , d. w. hogg , i. c. v. ivezi , g. r. knapp , d. q. lamb , b. c. lee , r. h. lupton , t. a. mckay , p. kunszt , j. a. munn , l. oconnell , j. peoples , j. r. pier , m. richmond , c. rockosi , d. p. schneider , c. stoughton , d. l. tucker , d. e. vanden berk , b. yanny , and d. g. york _ phys .
d _ * 69 * , p. 103501
, may 2004 .
l. brink , b. cabrera , j. p. castle , j. cooley , l. novak , r. w. ogburn , m. pyle , j. ruderman , a. tomada , b. a. young , j. filippini , p. meunier , n. mirabolfathi , b. sadoulet , d. n. seitz , b. serfass , k. m. sundqvist , d. s. akerib , c. n. bailey , m. r. dragowsky , d. r. grant , r. hennings - yeomans , and r. w. schnee _ nuclear instruments and methods in physics research a _ * 559 * , pp . 414416 , apr .
2006 .
s. leman , b. cabrera , k. mccarthy , m. pyle , r. resch , b. sadoulet , k. sundqvist , p. brink , m. cherry , do couto e silva , e. , e. figueroa - feliciano , n. mirabolfathi , b. serfass , and a. tomada _ chinese journal of physics ( phonons 2010 conference proceedings ) _ * 49 * , p. 349
, 2011 . , k. mccarthy , p. brink , b. cabrera , m. cherry , e. do couto e silva , e. figueroa - feliciano , p. kim , n. mirabolfathi , m. pyle , r. resch , b. sadoulet , b. serfass , k. sundqvist , , a. tomada , , and b. young _ journal of applied physics _ * 110 * , p. 094515
, 2011 .
, s. hertel , p. kim , b. cabrera , e. do couto e silva , e. figueroa - feliciano , k. mccarthy , r. resch , b. sadoulet , k. sundqvist , and on behalf of the cryogenic dark matter search collaboration _ journal of low temperature physics _ * 167 * , pp . 11061111 , 2012 . ,
d. brandt , p. brink , b. cabrera , h. chagani , m. cherry , p. cushman , e. do couto e silva , t. doughty , e. figueroa - feliciano , v. mandic , k. mccarthy , n. mirabolfathi , m. pyle , a. reisetter , r. resch , b. sadoulet , b. serfass , k. sundqvist , a. tomada , b. young , j. zhang , and on behalf of the cryogenic dark matter search collaboration _ journal of low temperature physics _ * 167 * , pp .
10991105 , 2012 . | this review discusses detector physics and monte carlo techniques for cryogenic , radiation detectors that utilize combined phonon and ionization readout .
a general review of cryogenic phonon and charge transport is provided along with specific details of the cryogenic dark matter search detector instrumentation . in particular
this review covers quasidiffusive phonon transport , which includes phonon focusing , anharmonic decay and isotope scattering .
the interaction of phonons in the detector surface is discussed along with the downconversion of phonons in superconducting films .
the charge transport physics include a mass tensor which results from the crystal band structure and is modeled with a herring vogt transformation .
charge scattering processes involve the creation of neganov - luke phonons .
transition - edge - sensor ( tes ) simulations include a full electric circuit description and all thermal processes including joule heating , cooling to the substrate and thermal diffusion within the tes , the latter of which is necessary to model normal - superconducting phase separation .
relevant numerical constants are provided for these physical processes in germanium , silicon , aluminum and tungsten .
random number sampling methods including inverse cumulative distribution function ( cdf ) and rejection techniques are reviewed . to improve the efficiency of charge transport modeling , an additional second order inverse cdf method is developed here along with an efficient barycentric
coordinate sampling method of electric fields .
results are provided in a manner that is convenient for use in monte carlo and references are provided for validation of these models .
= 1 |
random ( and other ) paraxial optical fields generically contain numerous point topological singularities ( defects ) in a plane ( the @xmath14-plane ) oriented perpendicular to the propagation direction ( @xmath15-axis ) .
these singularities , which include phase singularities ( optical vortices ) @xcite polarization singularities ( c points ) @xcite , gradient singularities ( maxima , minima , and saddle points ) @xcite , and curvature singularities ( umbilic points ) @xcite , are the defining features of the optical field , and are characterized by signed winding numbers ( topological charges ) @xmath16 .
generically , vortices and gradient singularities have charge @xmath17 , whereas the charge of c points and umbilic points is @xmath18 . like electrostatic charges , topological charges screen one another :
positive charges tend to be surrounded by a net excess of negative charge , and vice versa hal81,liumaz92,robbod96,frewil98,berden00,den02,fresos02,fol03,den03,wil04,folgnu04,freego07,egosos07 .
a formal measure of screening is the charge correlation function @xmath19 @xcite , which by convention is constructed to measure the net excess of negative charge surrounding a positive charge at the origin .
when @xmath20 screening is said to be complete , whereas for partial screening @xmath21 . here
we concern ourselves with complete screening . in generic optical fields the average net charge @xmath22 , but there are large fluctuations : these are characterized in lowest order by their variance @xmath23 . for a purely random collection of @xmath24
singularities with say charges @xmath17 , @xmath25 .
screening significantly reduces these fluctuations , to an extent which depends on the rate of fall - off of @xmath26 for large @xmath27 . in the generic case where @xmath26 decays sufficiently rapidly , canonically exponentially , @xmath28 @xcite ; this is the hallmark of short range screening , . for a slowly decaying @xmath26
however , we shall see that the charge variance grows faster than @xmath29 ; we call this regime long range screening . @xmath0 and @xmath19 are closely related , and given the latter the former can , in principle , be calculated . here
, we explore in detail the behavior of @xmath0 and its relationship to @xmath19 for long range screening .
specifically , we examine a particular example , first suggested in @xcite , and obtain analytical results for the behavior of @xmath0 in a large circular region of radius @xmath2 , finding @xmath30 . in elliptically polarized paraxial fields
c points are vortices ( zeros ) of the right and left handed circularly polarized components of the field @xcite , so our results are applicable also to these singularities .
@xmath19 is also known for the stationary and the umbilic points of random circular gaussian fields , such as the real and imaginary parts of the optical field @xcite , but the correlation functions of these singularities differ from that of the vortices and our results are not necessarily applicable to these singularities .
measurement of @xmath19 ( either from experiment or computer simulation ) requires locating and characterizing all vortices in each of a large number of independent realizations - this is a challenging task . on the other hand , @xmath31 , and therefore @xmath32 ,
can , using the index theorem @xcite , be obtained from measurements made only on the boundary of the region of interest . as discussed in @xcite , a convenient way of accomplishing
this is to count signed zero crossings of either the real or imaginary parts of the optical field .
correlations between these zero crossings are described by the zero crossing correlation function @xmath33 . in @xcite
it was argued that like @xmath19 , also @xmath33 can be used to calculate @xmath0 ; this argument was supported by numerical results and computer simulations , but was not demonstrated analytically .
the plan of this paper is as follows : in section ii we discuss the connection between @xmath0 and @xmath19 in a natural , i.e. sharply bounded , region of finite area - first reviewing known results for short range screening , and then presenting what are to our knowledge the first explicit results for such boundaries for long range screening .
these latter are the major contribution of this report . in section iii
we discuss the connection between @xmath23 and the zero crossing correlation function @xmath11 , showing , analytically for the first time , the equivalence of the two seemingly different approaches to @xmath23 : one based on @xmath19 ; the other on @xmath10 . in section iv we consider a practical realization of an optical field with long range correlations that could be used to compare experiments with the results presented here . in the discussion in section
v we consider asymptotic forms of @xmath0 for very small , and very large , @xmath2 , for physically real optical fields .
we summarize our main findings in the concluding section vi .
in @xcite the following quantitative relationship between the charge variance @xmath0 and the charge correlation function @xmath34 was derived for a bounded region of area @xmath1 ( @xcite , eq .
( 37 ) ) with @xmath35 : @xmath36 here and throughout @xmath37 represents an ensemble average , @xmath24 is the number of singularities with charges @xmath38 contained in @xmath1 , @xmath39 is the average number density of singularities , and @xmath40 is the area of overlap between @xmath1 and its replica displaced by @xmath41 . in what follows
we assume isotropy , and therefore take @xmath1 to be a circular area with radius @xmath2 .
elementary geometry yields @xmath42 noting that @xmath43 , eq .
2 ] ) becomes @xmath44 + 2\pi \eta \int_{0}^{2r}rb(r)c(r)dr . \label{eq .
4}\ ] ] eq .
4 ] ) is our basic starting point .
we proceed to evaluate it for large @xmath2 : first for a characteristic field with short range correlations ; then for one whose correlations are long range .
halperin @xcite first showed that assuming stationarity and circular gaussian statistics , the charge correlation function for optical vortices can be written as @xmath45 } { \pi w^{\prime \prime } ( 0)r(1-w^{2}(r))^{2 } } , \label{eq .
5}\ ] ] where @xmath46 is the normalized autocorrelation function of the optical field @xmath47 , and @xmath48 , @xmath49 .
liu and mazenko @xcite , and berry and dennis @xcite , subsequently noted that eq .
( [ eq . 5 ] ) can be cast into the convenient form @xmath50 where @xmath51 the number density of vortices @xmath39 also depends on @xmath52 , and is @xcite @xmath53 a useful measure of how rapidly screening sets in is @xmath54 .
10b}\end{aligned}\ ] ] for short range screening @xmath55 @xmath56 for @xmath57 of order nearest neighbor separations , or greater ; for long range screening @xmath58 approaches @xmath56 only when @xmath59 .
if @xmath26 decays sufficiently rapidly with @xmath27 , the large @xmath27 contribution in the first integral in eq .
4 ] ) is negligible , and the upper limit can be extended to infinity . using the screening relationship in eq .
( [ eq . 1 ] ) we then have @xmath60 in this same limit @xmath61 in eq .
( [ eq . 3b ] ) can be replaced by its leading term in an expansion in powers of @xmath62 , @xmath63 , and we have @xmath64 where @xmath65 is the length of the perimeter of @xmath1 , and the screening length @xmath66 , which is a measure of the width of @xmath67 in eqs .
( @xmath68 ) , is @xcite @xmath69 thus , the hallmark of short range screening is that for @xmath70 , @xmath4 grows with the perimeter , rather than with the area , of @xmath1 .
these results have a simple physical explanation @xcite .
because charges deep inside @xmath1 are perfectly screened they make no contribution to the net charge @xmath31 , and therefore no contribution to its variance @xmath23 .
charges located within a distance of order @xmath71 from the boundary of @xmath1 , however , are imperfectly screened .
the number of these charges is @xmath72 . for completely unscreened charges @xmath73 , whereas for partially screened charges we can expect @xmath74 @xcite , where from eq .
12 ] ) @xmath75 .
12 ] ) and ( [ eq . lambdascrn ] ) also quantify the notion of short - range screening .
the condition that @xmath66 is finite is sufficient to insure that @xmath0 grows linearly with @xmath2 . thus , for short range screening @xmath26 ( averaged over a period in the case that it oscillates ) must fall faster than @xmath76 .
the canonical short range correlation function which gives rise to short range screening is the gaussian , which we write as @xmath77 where @xmath78 is the wavelength , @xmath79 is the ( asymptotically large ) distance between the random source and the screen on which the field @xmath47 is measured , and @xmath80 measures radial displacements on this screen .
@xmath81 is the normalized autocorrelation function of the far field speckle pattern produced by a distribution of randomly phased sources with amplitudes @xmath82 ^{2}\right ) , \label{eq .
15}\ ] ] where @xmath83 measures radial displacements in the source plane , and @xmath84 is a measure of the width of the distribution .
the displacement @xmath41 and the displacement @xmath85 are related by the scale factor @xmath86 @xmath87 giving for the gaussian @xmath88 then , using eq .
( [ eq . lambdascrn ] ) we obtain @xmath89 where @xmath90 is the riemann zeta function .
the average spacing @xmath91 between singularities is @xmath92 which exceeds the screening length @xmath66 by nearly a factor of two !
this apparent paradox is resolved in fig .
[ fig1 ] , where it can be seen that the singularities tend to cluster with nearest neighbor spacings of order @xmath66 .
to @xmath93 black to white for a field with ( a ) a gaussian and ( b ) a @xmath5 correlation function .
positive ( negative ) singularities are shown by filled white ( black ) circles .
both fields have the same average number density of singularities .
( a ) gaussian correlation function . here
there are @xmath94 positive and @xmath94 negative singularities .
the screening length @xmath66 is shown by the bar .
note the clustering of singularities into small , charge - neutral groups with the spacing between nearest neighbor singularities being of order @xmath66 , and with large empty regions between groups .
it is this clustering that produces a screening length @xmath95 the average spacing between singularities .
( b ) @xmath5 correlation function . here
there are @xmath96 positive and @xmath96 negative singularities .
as can be seen , there is a strong tendency for the singularities to be nearly equally spaced on a lattice - like structure in which each nearly square unit cell , shown by the dotted rectangle , contains two positive and two negative singularities .
this structure exhibits substantial local charge neutralization , but because of defects in the lattice the overall screening is long range .
, scaledwidth=70.0% ] a measure of how quickly screening sets in , thereby leading to local charge neutrality , is the rate of decay of @xmath97 eqs .
( 10 ) : for @xmath98 , @xmath99 @xmath100 , for @xmath101 , it equals @xmath102 , and for @xmath103 , @xmath104 has dropped below @xmath105 . in ( @xcite , eq .
( 4.48 ) and ( 4.49 ) ) a gauss - smoothed boundary is assumed , and a _ constant _ , i.e. @xmath2 independent , value for the charge variance , here @xmath106 , is obtained from @xmath107 , \label{eq .
55a } \\ & = & \frac{1}{4}\int_{0}^{\infty } x\frac{w^{\prime } \left ( x\right ) ^{2}}{% 1-w^{2}(x)}dx , \label{eq .
55b}\end{aligned}\ ] ] where @xmath9 in @xcite is our @xmath108 , @xmath109 with @xmath110 given in eq .
( [ eq . 5 ] ) , @xmath1 is the area of the region of interest , here @xmath111 , and @xmath112 @xmath113 is the total number of charges inside this region .
( [ eq . 55b ] ) is obtained from eq .
( [ eq . 55a ] ) by using eqs .
( [ eq . 1 ] ) , ( [ eq . 7 ] ) , ( 8) , and ( [ eq .
eta ] ) , expanding @xmath114 keeping the leading term in @xmath115 , and integrating by parts . for the gaussian correlation function in eq .
17 ] ) , eq .
55b ] ) can be evaluated analytically , and we obtain @xmath116 this result shows that @xmath117 is not only independent of @xmath2 , but also of @xmath84 , and is therefore independent of the charge density @xmath39 and all other system parameters ! a question of considerable interest not discussed in @xcite is whether or not this value for @xmath117 is an intrinsic property of the medium , or is it the result of a particular set of assumptions ? in @xcite it is stated that eq .
( [ eq . 55a ] ) is obtained from a calculation `` with the boundary of @xmath1 gauss - smoothed to eliminate trivial edge effects '' , but the details of the calculation are not given . in order to be able to more closely examine the physical content of eq .
55a ] ) we need a derivation of this equation .
here we briefly outline such a derivation , starting with the definition of a boundary that is `` gauss - smoothed '' .
@xmath118 in eq .
( 3 ) can be obtained from an area function @xmath119 using @xmath120 .
57}\ ] ] throughout this report we have used a disk shaped area described by the area function @xmath121 this function , shown in fig .
[ fig2 ] as curve ( a ) , inserted into eq .
57 ] ) yields @xmath122 which when evaluated yields eq .
( 3 ) . for different values of @xmath123 in eq .
curve a is the disk area function , eq .
58 ] ) , formally eq .
64 ] ) with @xmath124 .
curves b - e are eq .
64 ] ) with the following values for @xmath125 : curve b , @xmath126 , a so called super gaussian ; curve c , @xmath127 , a normal gaussian ; curve d , @xmath128 , a normal exponential ; curve e , @xmath129 , a so called stretched exponential.,scaledwidth=50.0% ] we note that in order to maintain the same number of charges @xmath24 within @xmath1 as there are for the disk , any other form for @xmath130 must satisfy @xmath131 for a gauss - smoothed boundary we have instead of eq .
59 ] ) , @xmath132 which is shown in fig .
2 as curve ( c ) . inserting eq .
61 ] ) into eq .
57 ] ) yields @xmath133 a peculiarity of eq .
62 ] ) is that @xmath134 , as opposed to @xmath135 , the expected result , which is obtained from eq .
( 3 ) . as a result of this peculiarity , in the absence of screening the gauss - smoothed calculation yields @xmath136 , instead of the expected random walk result @xmath137 .
generalizing eq .
4 ] ) to an arbitrary area function satisfying eq .
( [ eq . 60 ] ) , we have @xmath138 inserting eq .
( [ eq . 62 ] ) , and using @xmath139 we recover eq .
( [ eq . 55a ] ) .
now , if the physical content of @xmath117 were that it is an intrinsic property of the random medium , rather than simply the result of a calculation for a particular form for @xmath140 , then other forms for @xmath130 should yield the same result . in order to test this we redo the above calculation using a boundary smoothed by a general exponential function of the form , @xmath141 where @xmath142 is a real , not necessarily integer , number , and in order to satisfy eq .
( [ eq . 60 ] ) , @xmath143 ^{\left ( p/2\right ) } .
\label{eq . 65}\ ] ] examples of @xmath130 for different values of @xmath123 are shown in fig .
2 as curves ( b ) , ( d ) , and ( e ) . inserting eqs .
64 ] ) and ( [ eq .
65 ] ) into ( [ eq . 57 ] ) , we find @xmath144 so that in the absence of screening @xmath145 , which approaches @xmath24 for large @xmath123 , and approaches zero as @xmath123 approaches zero . for the charge variance itself
we have @xmath146 which reduces to the gaussian result in eq .
56 ] ) for @xmath127 , but differs from it for other values of @xmath123 , demonstrating that @xmath147 is , in fact , not an intrinsic property of the medium , but is dependent on the particular choice of @xmath148 . in @xcite a short range autocorrelation function of the form @xmath149 was used .
for this form for arbitrary @xmath123 we obtain by numerical integration @xmath150 which is again independent of @xmath84 . because both eqs .
67 ] ) and ( [ eq .
69 ] ) are continuous functions of @xmath123 , a particular value for @xmath117 does not uniquely specify neither the field autocorrelation function nor the boundary smoothing function .
for example , @xmath151 for a gaussian autocorrelation function and a gaussian boundary smoothing function for which @xmath127 .
but the exact same value for @xmath117 is obtained using the @xmath152 autocorrelation function in eqs .
( [ eq . 68 ] ) and a boundary smoothing function with @xmath153 . berry and dennis @xcite
have introduced the long range correlation function @xmath154 where @xmath155 is a bessel function of integer order @xmath156 .
this form for @xmath157 arises from a random source distribution @xmath158 that takes the form of a very thin ring , @xmath159 practical realization of a @xmath5 correlation function is discussed in section iv . using eqs .
( 10 ) with @xmath59 the screening relationship in eq .
( [ eq . 1 ] ) is easily verified , whereas upon inserting eqs .
( [ eq . 7 ] ) and ( [ eq . omega ] ) into eq .
( [ eq . lambdascrn ] ) it is immediately seen that @xmath66 does not converge .
these are the conditions for long range screening . inserting eqs .
( [ eq . 3b ] ) , ( [ eq . 7 ] ) , and ( [ eq . omega ] ) into eq .
4 ] ) , and integrating by parts , we obtain @xmath160 in the remainder of this report we use this highly convenient form for @xmath161 exclusively . for the case of @xmath5 , eq .
20 ] ) , we break the integral on the r.h.s . of eq .
( [ eq . 22 ] ) into two parts : @xmath162 where @xmath163 with @xmath164 . in evaluating @xmath165 and @xmath166 we will need the asymptotic forms @xmath167 expanding @xmath168 and keeping only the leading @xmath2 term we have for @xmath169@xmath170 this form diverges as @xmath171 , and
so we regularize it by writing @xmath172 , \label{eq .
27a } \\ & \approx & \frac{ar}{\pi } \left [ { \cal d}+\int_{0}^{a\delta } dx\,j_{1}^{2}(x)% \right ] , \label{eq .
27b}\end{aligned}\ ] ] where @xmath173 is evaluated numerically . using eq .
( [ eq . 25 ] ) the remaining integral in eq .
( [ eq . 27b ] ) can be evaluated analytically in the limit of large @xmath174 as follows : @xmath175 , \label{eq .
27bx } \\ & \approx & \lim_{\epsilon \rightarrow 0^{+}}\left [ \frac{\gamma \left ( \frac{% 3+\epsilon } { 2}\right ) } { 2^{\epsilon } \gamma ( 2)\gamma \left ( \frac{% 1+\epsilon } { 2}\right ) } { } _ { 2}f_{1}\left ( \frac{3-\epsilon } { 2},\frac{% 1-\epsilon } { 2};2;1\right ) -\int_{a\delta } ^{\infty } x^{-\epsilon } \frac{1}{% \pi x}\right ] , \label{eq .
27cx } \\ & = & \lim_{\epsilon \rightarrow 0^{+}}\left [ \frac{\gamma \left ( \frac{% 3-\epsilon } { 2}\right ) } { 2^{\epsilon } \gamma ( 2)\gamma \left ( \frac{% 1+\epsilon } { 2}\right ) } \frac{\gamma ( 2)\gamma ( \epsilon ) } { \gamma \left ( \frac{1+\epsilon } { 2}\right ) \gamma \left ( \frac{3+\epsilon } { 2}\right ) } -% \frac{1}{\epsilon \pi } ( a\delta ) ^{-\epsilon } \right ] , \label{eq .
27dx } \\ & = & \frac{\gamma + 3\ln 2 - 2}{\pi } + \frac{1}{\pi } \ln ( a\delta ) , \label{eq .
27ex}\end{aligned}\ ] ] where @xmath176 is euler s constant .
similarly , in calculating @xmath166 , eq .
( [ eq . 24bx ] ) , we use the asymptotic form in eq .
25 ] ) to obtain @xmath177 _ { \delta } ^{2r } , \label{eq . 29bx } \\ &
\approx & \frac{a}{2\pi ^{2}}\left [ -2r\ln \frac{4r^{2}}{2r}-2r+2r\ln \frac{% 8r^{2}}{\delta } \right ] , \label{eq .
29cx } \\ & = & \frac{ar}{\pi ^{2}}\left [ \ln \frac{4r}{\delta } -1\right ] .
29dx}\end{aligned}\ ] ] combining @xmath165 and @xmath166 , we obtain the central result of this report , @xmath178 } $ } , \label{eq . 30bx}\ ] ] where @xmath179 we note that for rather modest values @xmath180 the @xmath181 term dominates , so that this term should be easily accessible to experiment . direct numerical integration of eq .
2 ] ) reveals that eq .
( [ eq . 30bx ] ) is good to better than @xmath182 for @xmath183 , to better than @xmath184 for @xmath185 , and that thereafter the percentage error decreases as @xmath186 . as for the gaussian ,
a measure of how quickly ( or slowly ) screening sets in is the average rate of decay of @xmath187 . for large @xmath188
this average rate is @xmath189 , showing that although screening is ultimately complete , it approaches completeness very slowly .
what does boundary smoothing do to @xmath0 when the correlations are long ranged ? for the long range correlation function @xmath190 , @xcite states
`` we can show from ( 4.48 ) [ here eq .
( [ eq . 55a ] ) ] that ...
@xmath191 '' , i.e. for a circular region of radius @xmath2 , @xmath192 . for large @xmath2 ,
the limit also used in @xcite , we find here for all generalized exponential smoothing functions in eq .
@xmath193 for the gaussian smoothing function used in @xcite we obtain analytically @xmath194 in general , however , @xmath195 must be evaluated numerically . introducing scaled variables @xmath196 , and using the asymptotic form @xmath197 , eq .
( [ eq . 25 ] ) , we have @xmath198 carrying out the required numerical integrations we obtain : @xmath199 ; @xmath200 ; @xmath201 ; and @xmath202 . thus , also for long range correlations the results of boundary smoothing change when the arbitrary smoothing function is changed .
as indicated in the introduction , @xmath31 , and therefore @xmath23 , can be obtained from measurements made only on the boundary of the region of interest @xmath1 , here a circle of radius @xmath2 .
the method suggested in @xcite , which is amenable to calculation , is the following : signed zero crossings ( zcs ) of either the real @xmath204 , or of the imaginary @xmath205 , parts of the wave functions that cross the boundary are counted , with each positive ( negative ) zc contributing @xmath206 ( @xmath207 ) to @xmath31 ; the sign @xmath208 of a zero crossing of @xmath204 is @xmath209 sign@xmath210 , that of a zc of @xmath211 is @xmath212 sign@xmath213 , where @xmath214 , @xmath215 @xcite .
writing the lineal number density of positive ( negative ) zero crossings of @xmath216 or of @xmath205 as @xmath217,(@xmath218 ) , their sum and difference are @xmath219 where , for circular gaussian statistics @xcite @xmath220 for a straight line of length @xmath221 oriented along say the @xmath115-axis @xcite , @xmath222 . \label{eq . 33x}\ ] ] the zero crossing correlation function @xmath203 is @xmath223 where @xmath224 is obtained from eq .
( [ eq . omega ] ) with @xmath27 replaced by @xmath225 .
@xmath226 and @xmath9 are compared in fig .
[ fig3 ] . and
( b ) zero crossing correlation function @xmath227 for short range gaussian field correlator @xmath228 .
( c , d ) long range correlations .
( c ) charge correlation function @xmath229 and ( d ) zero crossing correlation function @xmath227 for long range field correlator @xmath230 .
short range correlations : @xmath229 in ( a ) is everywhere negative , indicating that on average positive ( negative ) charges are surrounded by a tight cloud of excess negative ( positive ) charge , leading to local charge neutralization - this is the characteristic field structure of short range screening , see fig .
similarly , @xmath231 in ( b ) is also everywhere negative , indicating that a positive ( negative ) zero crossing on the boundary is likely to be followed by an excess of negative ( positive ) zero crossings that quickly cancel its contribution to the net charge .
this structure leads to a linear increase of the charge variance with boundary length .
long range correlations : although for , say , a central positive charge the first negative dip in @xmath229 in ( c ) integrates to an excess negative charge of @xmath232 , indicating substantial , albeit incomplete , local charge neutralization , much of this is offset by the following positive peak . as can be seen , the net average field structure consists of rings whose weak excess charge alternates positive / negative , with each successive negative ring having marginally greater area , and therefore marginally greater negative charge , than the positive ring that follows it .
these oscillating cancellations give rise to a slow @xmath233 approach to complete screening that is the hallmark of long range screening .
this weak structure is too diffuse to be seen in fig .
for the zero crossing correlation function @xmath231 in ( d ) , a positive ( negative ) zero crossing on the boundary is likely to be followed by a long string of zero crossings with a weak excess of positive ( negative ) signs that reflect the bulk ring structure ( the boundary runs through one of these rings ) .
this structure leads to an @xmath234 growth of the charge variance .
, scaledwidth=70.0% ] for a circle of radius @xmath2 , @xmath235 , \label{eq .
35x}\end{aligned}\ ] ] where @xmath236 measures the angular separation of two points on the rim of the circle .
this form for @xmath237 follows from section f of @xcite with @xmath115 replaced by @xmath238 .
the corresponding form for @xmath239 is @xmath240 where @xmath241 is the number of zero crossings per unit of arc that cross the rim of the circle , and @xmath242 is obtained from eq .
( [ eq . 34bx ] ) with @xmath243 replaced by @xmath244 .
this form for @xmath239 follows from section g of @xcite with @xmath115 replaced by @xmath245 . using the fact that in the limit @xmath236 goes to zero the difference between the rim ( arc ) of the circle @xmath246 and the corresponding chord@xmath247 here @xmath27 , vanishes , together with @xmath248 , we have @xmath249 where @xmath250 , given in eq .
( [ eq . 31ax ] ) , is the lineal number density of zero crossings . in what follows
we refer to eq .
( [ eq . 33x ] ) as the _ linear _ @xmath203 formula , and to eq .
( [ eq . 35x ] ) as the _ circular _ @xmath203 formula .
the field autocorrelation function @xmath251 appearing in @xmath242 depends only on the straight line separation , i.e. chord length @xmath27 , between two points on the circle rim separated by @xmath236 , @xmath252 we now proceed to show explicitly that @xmath0 in eq .
[ eq . 2 ] , which is written in terms of the charge correlation function @xmath253 and the area of @xmath1 , and @xmath0 in the circular @xmath203 formula , eq .
( [ eq . 35x ] ) , which is written in terms of the zero crossing correlation function @xmath203 and the perimeter of @xmath1 , are identical - as they must be . returning to eq . ( [ eq .
22 ] ) we write @xmath254 \right\ } . \label{eq . 39bx}\end{aligned}\ ] ] we now compare this with the circular @xmath203 formula , eq .
( [ eq . 35x ] ) .
writing @xmath255 , \label{eq .
40x}\ ] ] and using the fact that @xmath256 , so that @xmath257 , we have @xmath258 , \label{eq .
41ax } \\ & = & \frac{n_{o}\pi r}{2}-\frac{1}{2\pi } \int_{0}^{2r}dr\sin ^{-1}w(r)\frac{d}{% dr}\left [ \sqrt{4r^{2}-r^{2}}\frac{w^{\prime } ( r)}{\sqrt{1-w^{2}(r)}}\right ] , \label{eq .
41bx}\end{aligned}\ ] ] _ q.e.d .
_ it was suggested in @xcite that the zero crossings of @xmath204 or of @xmath259 need not necessarily be counted over the actual perimeter @xmath8 of @xmath260 , but rather that in an isotropic system one could perform the count over any straight line with the same length as @xmath8 .
this suggestion , supported by numerical analysis and computer simulations , was based on the notion that for short range correlations the geometry of the line - rectangular , circular , straight , etc . -
was unimportant , as long as over most of its length its radius of curvature was large compared to the screening length @xmath261 . here
we demonstrate the validity of this suggestion analytically for short range correlations , and examine its validity for long range correlations .
we define an error parameter @xmath262 by @xmath263 where @xmath264 @xmath265 @xmath266 .
45x}\ ] ] for large @xmath2 , when @xmath267 is short ranged @xmath268 decays to zero well before the upper limit of the integral on the r.h.s .
( [ eq . 43x ] ) is reached . extending this limit to @xmath269
, it is easily seen that @xmath270 is a constant that does not grow with @xmath2 , and can therefore be neglected ; for the same reasons , @xmath271 in eq .
( [ eq . 44x ] ) can similarly be neglected .
expanding the square root in the r.h.s .
( [ eq . 45x ] ) , it is easily seen that also @xmath272 does not grow with @xmath2 , so it too can be neglected .
thus for short range correlations the linear @xmath203 formula and the exact result for @xmath273 are equivalent in the limit of large @xmath2 .
this equivalence can be demonstrated more directly .
returning to eq .
( [ eq . 22 ] ) , we have to leading order in @xmath62 @xmath274 , \label{eq , 46cx}\end{aligned}\ ] ] which is also the leading , i.e. @xmath2 dependent , term in the linear @xmath203 formula , eq .
( [ eq . 33x])@xmath275 we therefore conclude , in full accord with @xcite , that for short range screening the linear @xmath203 formula can , indeed , be used to measure @xmath161 . for long range correlations
we need the asymptotic form of @xmath268 , which for @xmath5 is @xmath276 to leading order the oscillating @xmath277 term makes no contribution , and we have : for @xmath278 , @xmath279 for @xmath271 , @xmath280 and for @xmath281 @xmath282 where in evaluating eq .
( [ eq . 50ax ] ) we use the asymptotic form @xmath283 assembling the pieces yields @xmath284 thus , unlike the case of short range screening , for long range screening the linear @xmath203 formula is always larger than the true result ; the percentage error , however , is not severe , being @xmath285 for @xmath185 , @xmath286 for @xmath287 , and @xmath288 for @xmath289 .
@xmath290 exceeds @xmath161 because setting @xmath291 in the integrals over @xmath292 in eq .
( [ eq . 33x ] ) implies that the correlations between zero crossings act along the arc of the circle .
but , as already noted in connection with the circular @xmath203 formula , these correlations act only along the chords of the circle
. however , except for the diameter , all chords are shorter than their arcs , so that @xmath291 necessarily overestimates the correlation effects .
we now inquire as to what value of @xmath221 in the upper limits of the correlation integrals appearing on the r.h.s . of eq .
( [ eq . 33x ] ) establishes equality between @xmath293 and @xmath273 .
writing @xmath294 , we have @xmath295 setting @xmath296 and solving for @xmath123 we obtain @xmath297 where lw is the lambertw function @xcite , and @xmath298 is the base of the natural logarithm .
53 ] ) suggests that as a physically attractive approximation @xmath221 should be set equal to the longest chord , @xmath299 , rather than to the longest arc , @xmath300 .
indeed , inserting @xmath128 into @xmath301 yields the excellent approximation @xmath302 - a forty - fold improvement over eq .
( [ eq . 51x ] ) .
as discussed in section iic , the long range correlation function @xmath304 arises from a source @xmath305 that takes the form of a ring of zero width , eq .
21 ] ) . here
we consider a ring of mean radius @xmath84 and finite width @xmath306 as a practical realization , @xmath307 , of @xmath308 .
we obtain @xmath307 from the van cittert - zernike theorem @xcite as @xmath309 , \label{eq .
54a } \\ & \approx & j_{0}\left ( ar\right ) \left [ 1-\frac{1}{6}\left ( \frac{b}{2a}% \right ) ^{2}\left ( ar\right ) ^{2}\right ] -\frac{1}{6}\left ( \frac{b}{2a}% \right ) ^{2}\left ( ar\right ) j_{1}\left ( ar\right ) + { \cal o}\left ( \left ( \frac{b}{2a}\right ) ^{4}\right ) .
\label{eq . 54b}\end{aligned}\ ] ] @xmath307 closely matches @xmath310 over the region @xmath311 , see fig .
[ fig4](a ) .
we can therefore expect for @xmath312 , and in fact do obtain , see figs .
[ fig4](b , c ) , a close match between the result obtained using @xmath313 in eqs .
( [ eq . 7 ] ) and ( [ eq . omega ] ) and integrating eq .
( [ eq . 22 ] ) numerically , and that obtained using @xmath314 . .
( a ) correlation functions .
solid curve @xmath230 , small circles @xmath315 equal to the practical realization of @xmath5 , eq . ( [ eq .
54a ] ) ; in both cases @xmath316 and @xmath317 .
the agreement between the two forms is excellent out to @xmath318 .
( b ) @xmath4 vs. @xmath319 .
solid curve the exact result , eq . ( [ eq . 22 ] ) with @xmath5 correlation function , small circles eq .
( [ eq . 22 ] ) with practical realization of @xmath5 .
the inset shows the percentage error obtained using the practical realization .
this error is less than @xmath320 for @xmath321 , and less than @xmath322 for @xmath287 .
, scaledwidth=70.0% ] a ratio @xmath323 appears to be experimentally practical , so that the long range screening results presented here should be amenable to experimentation .
what does smoothing the boundary of the wavefield mean both physically and mathematically ? close examination of the derivation of eq .
( [ eq . 55a ] ) reveals that _ all _ boundary smoothing functions act as filters that smoothly reduce to zero _ not _ the field amplitude as one recedes from the central point , but rather the _ number density of singularities_. how can such a fall - off be achieved ?
a physically meaningful possibility is to devise a source function @xmath324 that leads to a wavefield with an intrinsic singularity density that falls smoothly to zero from some central point . however , not only is there no suitable source function known , even in principle , but also such a wavefield , if it could be created , would violate a fundamental assumption in the derivation of the charge correlation function @xmath9 and in the calculation of @xmath0 that is used in @xcite as well as here , namely that the system is stationary , i.e. that on average it is the same everywhere , that there is no central point .
thus , for such a system the results given in @xcite , and our extension of them , need not necessarily apply .
so , if an intrinsic average number density of singularities that is independent of position is needed , then how can the fall - off required by the smooth boundary in @xcite be obtained ?
the only possibility that comes to mind is the following : some arbitrary point , the central point , is surrounded by a small circle and a series of narrow concentric rings .
suitable fractions of the singularities in these rings are then discarded to create a singularity density profile that approximates , say , the gaussian in @xcite , or any other desired smoothing function .
of course , an experimentalist would object that not only is such a procedure entirely arbitrary , but also that no good could ever come from throwing away data . and ,
as shown above , she would be right , because the cost of boundary smoothing for short range correlations is a loss of all wavefield information .
similarly , for long range correlations the magnitude of the calculated fluctuations change with changes in smoothing function , showing that the coefficient @xmath325 in eq .
( [ eq . dum_1 ] ) can not be interpreted as a screening length , as is done in eq .
12 ] ) .
surprisingly , perhaps , for sufficiently small @xmath2 , @xmath12 for all field correlation functions for which @xmath326 ; this includes not only the gaussian in eq .
17 ] ) , @xmath5 in eq .
20 ] ) , the practical approximation to @xmath5 in eq .
( [ eq . 54a ] ) , but also every other physically realizable short and long range correlation function .
@xmath326 follows from expansion of the hankel transform in the van cittert - zernike theorem @xcite , and the fact that all physically realizable @xmath324 are bounded .
the proof of this universal @xmath3 law for @xmath327 is as follows : returning to eq .
( [ eq . 22 ] ) we have for @xmath328 , @xmath329 the physical meaning of this result is that for sufficiently small @xmath330 there can no be screening because the probability of finding the required negative charge within @xmath1 is vanishingly small .
( _ i _ ) if one had a form for @xmath108 for which @xmath331 decays asymptotically as @xmath332 with @xmath333 , then @xmath334 ( _ ii _ ) for @xmath335 , @xmath336 , eq . ( [ eq . 30bx ] ) .
( _ iii _ ) for @xmath337 , @xmath338 , eq . ( [ eq . 12 ] ) .
the proofs of assertions ( _ i _ ) and ( _ iii _ ) are as follows : for @xmath337 , eq .
lambdascrn ] ) converges , and therefore eq . ( [ eq . 12 ] ) holds .
for @xmath339 we rewrite eq .
( [ eq . 22 ] ) as @xmath340 where @xmath224 is given in eq .
( 8) , and @xmath341 .
for @xmath342 we have @xmath343 , and since this is subdominant , depending only linearly on @xmath2 , we discard it without further discussion . inserting the asymptotic form @xmath344 into @xmath345 , and writing @xmath346 ,
we have @xmath347 passing to the limit @xmath348 we obtain eq .
71 ] ) .
how small can @xmath349 be in practice ?
the answer is that asymptotically , i.e. for sufficiently large @xmath2 , @xmath350 .
this is demonstrated below using the van cittert - zernike theorem @xcite and the fact that all physically realizable sources @xmath351 must be : ( _ i _ ) nonsingular , a requirement that excludes , inter alia , delta functions ; ( _ ii _ ) positive definite ; and ( _ iii _ ) strictly bounded , i.e. @xmath352 . in physically realizable sources whose finite extent is defined by a mask where the intensity falls discontinuously from some non - negligible value to zero at the mask edge @xmath353 , we have for large @xmath27@xmath354 so that both @xmath52 and @xmath355 decay as @xmath356 , i.e. @xmath357 .
if @xmath358 is extremely small the @xmath356 decay of @xmath359 sets in at values of @xmath27 that may be so large as to be unmeasurable in practice . in that case , in practice one observes @xmath360 to show this we set @xmath361 equal to infinity and write @xmath362 where @xmath363 . in @xmath364we
expand @xmath365 and obtain for large @xmath27 @xmath366 \nonumber \\ & & + s^{\prime } ( 0)\left [ -\sqrt{\frac{2}{\pi } } \frac{\delta ^{3/2}\cos ( r\delta + \pi /4)}{r^{3/2}}+\frac{11}{8}\sqrt{\frac{2}{\pi } } \frac{\delta ^{1/2}\sin ( r\delta + \pi /4)}{r^{5/2}}-\frac{1}{r^{3}}\right ] .
75}\end{aligned}\ ] ] using the large argument expansion of @xmath5 in @xmath367 and integrating by parts we have @xmath368 \,du \nonumber \\ & \approx & \sqrt{\frac{2}{\pi } } \frac{\cos ( r\delta + \pi /4)}{r^{3/2}}\left [ s(0)+\delta s^{\prime } ( 0)\right ] \delta ^{1/2 } \label{eq .
76a } \\ & & -\frac{\sin ( r\delta + \pi /4)}{r^{5/2}}\left ( \frac{s(0)+\delta s^{\prime } ( 0)}{2\delta ^{1/2}}+s^{\prime } ( 0)\delta ^{1/2}\right ) \nonumber \\ & & + \frac{1}{8}\frac{\sin ( r\delta + \pi /4)}{r^{5/2}}\left [ s(0)+\delta s^{\prime } ( 0)\right ] \delta ^{-1/2 } \label{eq .
76b } \\ & = & s(0)\sqrt{\frac{2}{\pi } } \left [ \cos ( r\delta + \pi /4)r^{-3/2}\delta ^{1/2}+\frac{3}{8}\sin ( r\delta + \pi /4)r^{-5/2}\delta ^{-1/2}\right ] \nonumber \\ & & + s^{\prime } ( 0)\sqrt{\frac{2}{\pi } } \left [ \cos ( r\delta + \pi /4)r^{-3/2}\delta ^{3/2}-\frac{11}{8}\sin ( r\delta + \pi /4)r^{-5/2}\delta ^{1/2}\right ] \label{eq . 76c}\end{aligned}\ ] ] summing @xmath364 and @xmath367 yields @xmath369 i.e. @xmath370 .
if , in addition , @xmath371 , the decay of @xmath52 is faster yet .
for example , when all odd derivatives of @xmath372 vanish at the origin , as happens for a gaussian @xmath351 , @xmath52 decays exponentially fast .
thus , for all physically realizable optical fields @xmath23 is bounded at its endpoints by two universal laws : for small @xmath2 , @xmath373 ; for large @xmath2 , @xmath374 .
this behavior is illustrated in fig .
5 for the practical @xmath5 autocorrelation function in section iv . , eq .
( [ eq . 54a ] ) with @xmath375 ; thin curve , exact @xmath5 , eq .
20 ] ) with @xmath316 .
both functions obey the universal small @xmath2 law , @xmath12 , eq .
( @xmath376 ) , as they must , but , as expected , only the physically realizable practical @xmath5 obeys the large @xmath2 law , @xmath374 , eq .
( [ eq . 12]).,scaledwidth=50.0% ]
under the assumption of circular gaussian statistics , the topological charge variance @xmath0 of vortices in scalar fields @xcite and c points in vector fields @xcite was analyzed for a circular region of area @xmath1 and radius @xmath2 for both short and long range correlation functions .
( _ ii _ ) for short range correlations @xmath0 grows linearly with @xmath2 , eq .
12 ] ) : due to screening , inside @xmath1 there is local charge neutralization , and only partially screened charges near the boundary contribute to the fluctuations .
very recent experimental measurements of @xmath0 @xcite have used eq .
12 ] ) to obtain the screening length @xmath66 , a fundamental wavefield parameter .
comparison of @xmath66 with the mean spacing between charges led to the conclusion that the charges form small clusters - a conclusion verified by direct imaging of the wavefield , fig . 1 . (
_ iv _ ) for long range correlation functions
@xmath23 grows faster than @xmath2 , and it is not possible to define a screening length . for a @xmath5 correlation function @xmath30 , eq .
( [ eq . 30bx ] ) . for correlation functions that decay more slowly than @xmath5 , @xmath378 , where @xmath379 .
( _ v _ ) although a @xmath5 correlation function is not attainable in practice , it was shown that one can generate an excellent approximation valid over an arbitrarily large , but finite range of @xmath2 , eq . ( [ eq . 54a ] ) ; this approximation was shown to yield results for @xmath23 that are in close agreement with those for @xmath5 , fig .
( _ vi _ ) @xmath0 can be calculated using either the charge correlation function @xmath9 , eqs .
( [ eq . 7 ] ) and ( @xmath380 ) , or the zero crossing correlation function @xmath33 , eq .
( @xmath381 ) .
an exact calculation showed that , as expected , these two seemingly different methods yield the same result for both short and long range screening .
( _ vii _ ) for short range screening it was shown that it is also possible to obtain @xmath0 from zero crossing measurements made along any straight line of length @xmath65 , rather than only along the circular perimeter of @xmath1 . for long range screening
, however , it was shown that this useful simplification no longer holds . | screening of topological charges ( singularities ) is discussed for paraxial optical fields with short and with long range correlations . for short range screening the charge variance @xmath0 in a circular region @xmath1 with
radius @xmath2 grows linearly with @xmath2 , instead of with @xmath3 as expected in the absence of screening ; for long range screening @xmath4 grows faster than @xmath2 : for a field whose autocorrelation function is the zero order bessel function @xmath5 , @xmath6 .
a @xmath5 correlation function is not attainable in practice , but we show how to generate an optical field whose correlation function closely approximates this form ; screening in such a field is well described by our theoretical results for @xmath7 .
@xmath0 can be measured by counting positive and negative singularities inside @xmath1 , or more easily by counting signed zero crossings on the perimeter @xmath8 of @xmath1 .
for the first method @xmath4 is calculated by integration over the charge correlation function @xmath9 , for the second by integration over the zero crossing correlation function @xmath10 . using the explicit forms of @xmath9 and of @xmath11 we show that both methods of calculation yield the same result .
we show that for short range screening the zero crossings can be counted along a straight line whose length equals @xmath8 , but that for long range screening this simplification no longer holds .
we also show that for realizable optical fields , for sufficiently small @xmath2 , @xmath12 , whereas for sufficiently large @xmath2 , @xmath13 these universal laws are applicable to both short and pseudo - long range correlation functions . |
although pulsar - like stars have many different manifestations , they are populated mainly by rotation - powered radio pulsars .
a lot of information about pulsar radiative process is inferred from the integrated and individual pulses , the sub - pulses , and even the micro - structures of radio pulses . among the magnetospheric emission models , the user - friendly nature of ruderman & sutherland ( 1975 ; hereafter rs75 ) model is a virtue not shared by others @xcite . in rs75 and its modified versions ( e.g. , * ? ? ?
* ) , a vacuum gap exists above polar cap of a pulsar , in which charged particles ( electrons and positrons ) are accelerated because of @xmath1 .
these accelerated charged particles , moving along the curved magnetic field lines , radiate curvature or inverse - compton - scattering - induced high energy photons which are converted to @xmath2 while propagating in strong magnetic field .
a follow - up breakdown of the vacuum gap produces secondary electron - positron pairs plasma that radiate coherent radio emission .
these models with gap - sparking provide a good framework to analyze observational phenomena , especially the drifting @xcite and bi - drifting @xcite sub - pulses . however , the rs75-like vacuum gap models work only in strict conditions : strong magnetic field and low temperature on surface of pulsars ( e.g. , * ? ? ?
* ; * ? ? ?
the necessary binding energy of positive ions ( e.g. , @xmath0fe ) for rs75 model to work should be higher than @xmath3 kev , while calculations showed that the cohesive energy of @xmath0fe at the neutron star surface is @xmath4 kev @xcite .
this binding energy problem could be solved within a partially screened inner gap model @xcite for normal neutron stars .
alternatively , it is noted that the binding energy could be sufficiently high if pulsars are bare strange quark stars @xcite although strange stars were previously supposed to exist with crusts @xcite .
certainly , it is very meaningful in the elementary strong interaction between quarks and the phases of cold quark matter that the binding energy problem could be solved by bare quark stars as pulsars @xcite .
though the ideas of solving the binding energy problem in bss model were presented and discussed in some literatures , there is no comprehensive study with quantitative calculations up to now . in this paper
, we are going to investigate the bss model in quantitative details and show the physical picture of binding of particles on bss s surface .
our research results are that multi - accelerators could occur above the polar cap for ( and only for ) the curvature - radiation - induced ( cr - induced ) sparking normal pulsars ( nps ) , but for other cases , such as resonant inverse - compton - scattering - induced ( ics - induced ) sparking nps and both cr - induced and ics - induced millisecond pulsars ( msps ) , particles on surface of bsss are bound strongly enough to form vacuum gap and rs75-like models work well if pulsars are bsss .
on a bss s surface , there are positively ( @xmath5-quarks ) and negatively ( @xmath6- and @xmath7-quarks and electrons ) charged particles .
quarks are confined by strong color interaction , whose binding energy could be considered as infinity when compared with the electromagnetic interaction , while electrons are bound by electromagnetic interaction . therefore , in this paper we focus on the binding of electrons .
let s discuss briefly the binding of electrons in the bss model at first .
on one hand , assuming the electric potential at the top of rs75 vacuum gap is the same as that of the interstellar medium , one could then have a potential barrier for electrons by integrating the gap electric field from top to bottom in the vacuum gap .
this potential barrier could then prevent electrons streaming into magnetosphere . on the other hand ,
electrons above the stellar surface of bss are described in the thomas - fermi model , in which the total energy of eletrons on fermi surface would be a constant , @xmath8 . in previous work ( e.g. alcock et al . 1986 ) , this constant is chosen to be zero , @xmath9 , because they did nt consider the effect of spinning bss with strong magnetic fields . due to the unipolar generator effect , potential drop between different magnetic field lines is set up from pole to equatorial plane .
this potential drop could result in different @xmath8 , at different polar angle , @xmath10 , and the total energy of electrons would then be obtained by choosing certain zero potential magnetic field line ( i.e. , at @xmath11 or @xmath12 in fig . [ antipulsar ] ) .
finally , comparing the total energy of electrons with the height of the potential barrier in vacuum gap , we can see whether eletrons can stream into magnetosphere freely or not .
the distribution of electrons in bsss is described in the thomas - fermi model @xcite . in this model ,
equilibrium of electrons in an external electric field assures that the total energy of each electron on fermi surface is a constant , @xmath8 . for the case of extremely relativistic degenerate electron gas
, it gives @xcite @xmath13 where @xmath14 is the total energy , @xmath15 is the fermi energy , @xmath16 is the electrostatic potential energy of electrons and @xmath17 is a constant , describing the potential energy of electrons in the thomas - fermi model at infinity . on the other hand , the potential distribution of electrons on the star s surface due to the electric field induced by the rotating , uniformly magnetized star , for the sake of simplicity , could be assumed and estimated as ( xu et al .
2006 , eq . 2 there )
@xmath18 where @xmath19 , and @xmath20 is the radius of a pulsar , @xmath21 is the pulsar period , @xmath10 is the polar angle and @xmath22 is another constant . in view of the distribution of electron above the surface of bss extends only thousands of femtometers , the macroscopic potential drop between different magnetic field lines could be thought to be at infinity in the thomas - fermi model . and the potential energy related to eq .
[ ugp ] , @xmath23 , could be regarded as the constant , @xmath17 , in eq .
[ fd ] . by choosing the certain zero potential magnetic field line ,
we could obtain the total energy of electrons , namely @xmath23 .
two scenarios could be possible here .
the first scenario is that we choose the critical field lines whose feet are at the same electric potential as the interstellar medium @xcite as the zero potential .
we may also suggest a second choice that the zero potential should be at those magnetic field lines which separate annular and core regions determined by @xmath24@xmath25 , where @xmath26 and @xmath25 , are the stellar surface areas of annular region and core region , respectively .
the second scenario is based on the idea that if particles with opposite charge stream into the magnetosphere with @xmath27 in both regions , areas of this two regions should approximately be equal in order to keep the star not charging .
the feet of the critical field lines and the magnetic field lines determined by @xmath24@xmath25 are designated as c and b , respectively ( fig .
[ antipulsar ] ) . for the above two scenarios ,
the total energy , @xmath28 , of electrons on the fermi surface are given by @xmath29 and @xmath30 respectively , where @xmath12 and @xmath11 are polar angles of c and b ( see fig .
[ antipulsar ] ) .
equations [ eea ] and [ eeb ] imply that the total energy of electrons is higher at the poles and decreases toward the equator for an antipular ( @xmath31 ) , which means that electrons in different regions above a polar cap may behave differently . in the following , we will consider the potential barrier of electrons in vacuum gap .
unlike rs75 , we do calculations in situation of an antipulsar whose magnetic axis is parallel to its spin axis . a schematic representation for antipulsar is shown in fig . [ antipulsar ] .
assuming the electric potential at the top of rs75 vacuum gap is the same as that of the interstellar medium , we could get a potential barrier for electrons by integrating the gap electric field from top to bottom in the vacuum gap .
this potential barrier , in one - dimensional approximation , is ( rs75 ) @xmath32 where @xmath33 is the height of vacuum gap , @xmath34 is the space coordinate measuring height above the quark surface .
this potential barrier may prevent electrons injecting into pulsar s magnetosphere .
the height of this potential barrier mainly depends on the height of vacuum gap which is determined by cascade mechanics of sparking , i.e. , the cr - induced cascade sparking and the ics - induced cascade sparking . in cr - induced cascade sparking model , the gap height is ( rs75 ) @xmath35 and in ics - induced cascade sparking model , it is @xcite @xmath36 in previous work of gil et al .
( 2006 ) , the heights of the vacuum gap of both cr - induced and ics - induced sparking mechanism ( gil et al .
2006 , eqs . 21 and 22 there )
are different from what we used in this work . in the psg model
, there was a partial flow of iron ions from the positively charged polar cap which coexist with the production of outflowing electron - positron plasmas .
such a charge - depleted acceleration region is also highly sensitive to both the critical ion temperature and the actual surface temperature of the polar cap @xcite .
differently , in our model , there is no flow of positively charged particles , namely quarks and also it is insensitive to the actual surface temperature .
this means that there is no partial screened effect above polar cap of bare strange quark stars , namely the pure vacuum gap exists on polar cap of bare strange quark stars .
that s the reason why we use eqs . [ hcr ] and [ hics ] in our calculation . whether this choice of height of vacuum gap could result in different driftrate of subpulses or not is a complicated problem
. we will discuss this problem very briefly in 3 .
the potential barrier of electrons in the gap for cr - induced cascade sparking model of typical normal pulsars ( nps ) is plotted in fig .
[ pb ] , in which the total energy of electrons at the stellar surface , namely @xmath37 , is illustrated at different polar angles .
the situation of cr - induced cascade sparking of typical millisecond pulsars ( msps ) is similar to that of nps but with greater height of potential barrier . , in vacuum gap of typical nps ( @xmath38 , @xmath39 g ) .
the potential energy of electrons at stellar surface , namely @xmath40 , is illustrated with fixed polar angles , for example , with @xmath41 and @xmath42 , where @xmath43 is the polar angle of the feet of the last open field lines ( fig .
[ antipulsar ] ) . ] comparing the potential barrier with total energy of electrons , we will explain behavior of electrons above polar cap .
namely , only electrons with energy greater than the potential barrier can escape into pulsar s magnetosphere .
it is known that energy of electrons is a function of polar angle ( eqs .
[ eea ] and [ eeb ] ) . as a result
, there may be a critical polar angle , @xmath44 , at which the energy of electrons equals the height of this potential barrier .
comparison between the total energy of electrons and the height of potential barrier on stellar surface for typical nps of cr - induced sparking is shown in fig .
[ pec ] ( @xmath44 does not exist for both ics - induced sparking of nps and msps , see table . [ pa ] ) .
the results are as follows : free flow status stays in the region of [ 0 , @xmath44 ] and vacuum gap in [ @xmath44 , @xmath43 ] for antipulsars , where @xmath43 is polar angle of the feet of the last open field lines ( fig .
[ antipulsar ] ) .
we give the results of @xmath44 in table .
[ pa ] for both pulsar and antipulsar , and find that for the special case of cr - induced sparking nps , free flow and vacuum gap could coexist above polar cap which differs from the previous scenario .
the general case is that only vacuum gap exists . and
@xmath47 , respectively .
the solid horizontal line is the height of the potential barrier of electrons , namely @xmath48 . ]
@lllllllll@ & & & & & & + + & @xmath43 ( rad ) & @xmath11 ( @xmath43 ) & @xmath12 ( @xmath43 ) & cr & ics & cr & ics & + & & & & 0.49 & ...
@xmath49 & 0.58 & ... & @xmath31 + nps & 0.0145 & 0.69 & 0.76 & 0.84 & 2.76@xmath50 & 0.90 & 2.83@xmath50 & @xmath51 + & & & & ... & ... & ... & ... & @xmath52 + msps & 0.145 & 0.69 & 0.76 & 1.49@xmath50 & ... & 1.52@xmath50 & ... & @xmath53 + + @xmath49@xmath44 does not exit , which means that the whole polar cap region is vacuum gap .
+ @xmath50@xmath44 @xmath54 @xmath43 , which means that the whole polar cap region is vacuum gap .
it follows from the previous argument that electrons inside bsss usually can not stream into magnetospheres .
does any other process which may affect the existence of vacuum gap above polar cap ? in vacuum gap , except pulling electrons from the interior of bsss , two other processes which may also prevent vacuum gap from being formed are required to be investigated .
one is the thermionic emission of electrons and another is the diffusion of electrons from the outer edge to the inner region of polar cap .
for the first one , if the current density due to thermionic emission of electrons is much smaller than that of goldreich - julian charge density , the vacuum gap could be maintained as well .
this current density is determined by the richard - dushman equation @xcite @xmath55 where @xmath56 is the mass of electron , @xmath57 is the boltzmann constant , @xmath58 is the temperature and @xmath59 is the work function of electrons . in the vacuum gap of bsss , the work function of thermionic electrons is the order of the difference between the height of the potential barrier and the total energy of electron at the surface of bsss .
the order of the difference is about @xmath60 mev . at the same time , the surface temperature of polar caps of bsss is order of @xmath60 k. thus , the thermionic emission current density is @xmath61 , which means that the thermionic emission of electrons can not affect the existence of the vacuum gap .
the second process is the diffusion of electrons whose distribution above bsss surface is @xcite @xmath62 eq .
[ en ] implies that the number density of electrons ( so does the kinetic energy density , @xmath63 ) decreases rapidly with increasing of the distance from quark matter surface at which @xmath63 @xmath64 @xmath65 , where @xmath66 is the magnetic field energy density . as a result , there is a balance surface where the kinetic energy density equals the magnetic energy density . below this balance surface
, electrons can cross magnetic field lines freely and above the balance surface , this motion is prevented . the physical picture of the diffusion of electrons is illustrated in fig .
[ df ] . making use of @xmath67 , where @xmath68 ( @xmath69 is the fermi energy of degenerate electrons ) and @xmath70
, we can obtain the height of the balance surface .
for nps , it is @xmath71@xmath72 and for msps , it is @xmath71@xmath73 , where @xmath74 .
keep in mind that there is a directed outward surface electric field above the quark matter surface .
this surface electric field is much stronger than the gap electric field but decreases rapidly with also increasing of the distance . which means that this surface electric field becomes smaller than the gap electric field above some certain distance , @xmath75 . for nps ,
it is @xmath75@xmath76 , and for msps , it is @xmath75@xmath77@xmath78 ( see fig . [ df ] ) .
both have @xmath71@xmath79 @xmath75 for nps and msps .
the diffusion of electrons beneath @xmath75 is still confined by the surface electric field meaning that only the diffusion of electrons above the surface with height of @xmath75 needs to be considered .
the diffusion coefficient , @xmath80 , is given by @xcite @xmath81 where @xmath82 ( @xmath83 is the larmor radius ) is the cyclotron radius of relativistic electrons , and @xmath84 is the mean free flight time of electrons .
the gradient of electrons along with the diffusion direction is approximately @xmath85 then , the diffusion rate is @xmath86 where @xmath87 .
for both the nps and msps with different @xmath88 , we give the results of the diffusion rate @xmath89 and @xmath90 in table [ drv ] in which the flow with the goldreich - julian flux is @xmath91 @xmath92@xmath93c@xmath94 @xmath77@xmath95@xmath96@xmath97@xmath98 @xmath99 @xmath100 . we can know that both have @xmath89 @xmath79 @xmath90 for nps and msps from table [ drv ] .
this means that the diffusion of electrons is also negligible which guarantees the existence of vacuum gap .
@ccccc@ & & + + @xmath88 ( mev ) & @xmath89 ( @xmath101 ) & @xmath90 ( @xmath102 ) & @xmath89 ( @xmath103 ) & @xmath90 ( @xmath104 ) + 1 & @xmath1054.75 & & @xmath1057.52 & + 10&@xmath1054.91 & @xmath105@xmath106 & @xmath1057.52 & @xmath105@xmath107 + 20&@xmath1054.93 & & @xmath1057.53 & +
in rs75 model , the binding energy problem is one of the most serious problems in the normal neutron star model of pulsars .
arons and scharlemann ( 1979 ) developed an alternative model , the space - charge limited flow ( sclf ) model , in which the particles , both iron ions and electrons can be pulled out freely , and form a steady flow @xcite . in this sclf model , the drifting sub - pulse phenomenon which has been commonly observed in pulsars can rarely be reproduced .
the prerequisite for understanding this phenomenon could be the existence of a vacuum gap . in a very special case , through our calculations , we find that there is a new physical scenario for cr - induced sparking of normal pulsars ( nps ) that free flow and vacuum gap may coexist above the polar cap .
but in other cases , such as ics - induced sparking of nps and millisecond pulsars ( msps ) , only vacuum gap exists . in general ,
if a pulsar is not highly negatively charged @xcite , vacuum gap survives at polar cap as well .
one limitation is that our calculation is based on one - dimensional approximation and it might fail in some cases of msps . as far as we find , it is very difficult to deal with the high - dimensional cases .
the one - dimensional approximation provides a good understanding of the geometry of polar cap of bsss . in conclusion
, the binding energy problem could be solved completely in the bss model of pulsar as long as bsss are neutral ( or not highly negative charged ) , and the structure of polar cap of bsss are very different with respect to that of nss .
detailed information about the geometry of bss s polar cap is given in table [ ag ] .
@llllll@ & & & + + & cr & ics & cr & ics & + & sclf & vg & vg & vg & @xmath31 + nps & vg & vg@xmath108 & sclf & vg@xmath108 & @xmath51 + & vg & vg & vg & vg & @xmath52 + msps & vg@xmath108 & vg & vg@xmath108 & vg & @xmath51 + + @xmath109@xmath44 represents @xmath110 while choosing @xmath111 and @xmath112 while choosing @xmath113 .
+ @xmath108for such cases , @xmath44 @xmath54 @xmath43 , which represents the structure of the whole polar cap region
. a more interesting region from pole to equator may locate between that polar angle where the total energy of electron equals the potential barrier and the polar angle of the foot of zero potential magnetic field line ( i.e. , [ @xmath114 or [ @xmath115 , see fig . [ pec ] ) for cr - induced sparking nps .
after the birth of a np , a vacuum gap exists at this region .
when sparking starts , the potential in vacuum gap drops rapidly due to screen by electron - positron pairs and may become lower than that at the surface , namely @xmath116 . as a result , the sparking converts vacuum gap to free flow at this region until the sparking ends , i.e. , at [ @xmath117 or [ @xmath115 , vacuum gap and free flow work alternately .
this argument may have profound implications for us to distinguish neutron stars and quark stars by pulsar s magnetospheric activities ( e.g. , the diversity pulse profiles ) .
another issue to be discussed is about the drifting rate of subpulses when we use the height of pure vacuum gap in this work .
the natural explanation of the drifting subpulse phenomena in vacuum gap is due to @xmath118 .
unfortunately , these theoretical calculations gave higher drifting rate with respect to observations ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
since it has been observed @xcite , the drifting subpulse phenomenon remains unclear which has been widely regarded as one of the most critical and potentially insightful aspects of pulsar emission @xcite .
the psg mechanism ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) could be a way to understand lower drifting rates observed , but some complexities still exist which make the underlying physics of drifting subpulses keep complicated and far from knowing clearly .
( 1 ) in principle the drifting velocity of subpulses is the ratio of the drifting distance to the duration , while the expected velocity predicted by @xmath118 is only for electrons in separated emission units , namely the plasma filaments
. these two velocities would not be the same if the plasma filaments may stop after sparking .
when sparking starts , the electric field in the vacuum gap vanishes due to screen by plasmas ; while sparking ends , the electric field appears again .
thus , the calculated drifting velocity with @xmath118 could be higher than that of observations .
( 2 ) the so - called aliasing effect : as one observes subpulses only once every rotation period , we can hardly determine their actual speed .
the main obstacles in the aliasing problem are the under sampling of subpulse motion and our inability to distinguish between subpulses especially when the differences between subpulses formed by various subbeams are smaller than the fluctuations in subpulses from one single subbeam @xcite .
anyway , detailed studies are very necessary in the future works .
we assume that the potential energy related to eq .
[ ugp ] , @xmath23 , to be the constant , @xmath17 , in eq .
this assumption could be reasonable .
for an uniformly magnetized , rotating conductor sphere , the unipolar generator will induce an electric field which is a function of polar angle , as described in eq .
[ ugp ] . in the case of @xmath119 ( fig .
[ antipulsar ] ) , the potential energy of electron is highest at the polar region which means that those electrons there could be easier to escape .
alternatively , this conclusion could be quantitatively understood as following : because of lorentz force inside a star , more electrons locate at the polar region so that the fermi energy of electron is higher there and easier to escape into magnetosphere .
we thank dr .
kejia lee and other members in the pulsar group of peking university for their helpful and enlightened discussions .
we also thank prof .
janusz gil for his helpful comments and suggestions .
junwei yu is grateful to dr .
caiyan li for her helpful assistance .
the work is supported by nsfc ( 10973002,10935001 ) , the national basic research program of china ( grant 2009cb824800 ) and the john templeton foundation .
melikidze , g. & gil , j. 2009 , the eighth pacific rim conference on stellar astrophysics ed b soonthornthum , s. komonjinda , cheng , k. s. & leung k. c. ( san francisco : astronomical society of the pacific ) 73 , 131 | in ruderman & sutherland ( rs75 ) model , the normal neutron stars as pulsars bear a severe problem , namely the binding energy problem that both ions ( e.g. , @xmath0fe ) and electrons on normal neutron star surface can be pulled out freely by the unipolar generator induced electric field so that sparking on polar cap can hardly occur . this problem could be solved within the partially screened gap ( psg ) model in the regime of neutron stars . however , in this paper we extensively study this problem in a bare strange quark star ( bss ) model .
we find that the huge potential barrier built by the electric field in the vacuum gap above polar cap could usually prevent electrons from streaming into the magnetosphere unless the electric potential of a pulsar is sufficiently lower than that at infinite interstellar medium .
other processes , such as the diffusion and thermionic emission of electrons have also been included here .
our conclusions are as follows : both positive and negative particles on a bss s surface would be bound strongly enough to form a vacuum gap above its polar cap as long as the bss is not charged ( or not highly negative charged ) , and multi - accelerators could occur in a bss s magnetosphere .
our results would be helpful to distinguish normal neutron stars and bare quark stars through pulsar s magnetospheric activities .
[ firstpage ] magnetospheric activities pulsars : general |
in the past decade , we have seen enormous progress in understanding flavor physics and @xmath1 violation . after turning on in 1999 , the new asymmetric - energy @xmath0 factories , pep - ii @xcite and kekb @xcite , quickly achieved luminosities that exceeded their design targets and the expectations of many .
this allowed the corresponding experiments , babar @xcite and belle @xcite , to quickly provide the first precision test of the ckm @xcite mechanism for @xmath1 violation .
the measurements of the proper - time - dependent @xmath1 asymmetry in charmonium-@xmath2 decays of neutral @xmath0 mesons ( @xmath3)@xcite are in very good agreement with the ckm prediction of @xmath3 from independent constraints .
it is convenient , both for visualization and quantitative analysis , to interpret experimental results within the ckm framework as constraints on the geometry of the so - called `` unitarity triangle '' @xcite , which is from the first and third columns of the ckm quark mixing matrix @xmath4 @xmath5 if one renormalizes the triangle by rescaling the sides by @xmath6 and adopts the wolfenstein phase convention @xcite , experimental results are interpreted as constraints on the apex of the triangle ( @xmath7 ) .
two independent groups ( ckmfitter@xcite and utfit@xcite ) provide the results of this analysis .
figure [ fig : rhoeta ] shows the constraints on the apex of the unitarity triangle as of the fpcp06 conference .
one can see that in addition to the precise determination of @xmath8 , mentioned above , the @xmath0 factory data strongly constrains the left side of the triangle , which is proportional to @xmath9@xcite .
the @xmath0 factory experiments have also measured the other two angles of triangle ( @xmath10@xcite and @xmath11@xcite ) . finally , the tevatron experiments , cdf and d0 , have recently measured @xmath12@xcite , the frequency of @xmath13 oscillations , which allows the right side of the triangle to be constrained , when combined with @xmath14 and some input from lattice qcd calculations@xcite .
it s clear from figure [ fig : rhoeta ] that the ckm model ( and thus the standard model ) gives a remarkably consistent description of all experimental results .
it s natural to ask whether there is still room for discovering physics beyond the standard model ( or new physics ) in @xmath0 decays , after seeing such consistency .
the answer is yes .
the most precise constraints in figure [ fig : rhoeta ] come from either tree - level @xmath0 decays or from @xmath0 mixing . from figure
[ fig : rhoeta ] , one can reasonably conclude that there is little room for substantial new physics corrections to tree - dominated processes @xcite . however , new physics may significantly alter the observables ( branching fractions , asymmetries , ... ) for rare @xmath0 decays without disturbing the beautiful consistency shown in figure [ fig : rhoeta ] .
rare @xmath0 decays are a unique and valuable tool in the search for new physics .
decays that are allowed at the tree level have relatively large amplitudes . if amplitudes from new physics are small , as we expect , hadronic uncertainties will prevent us from recognizing the presence of new physics in many cases . on the other hand ,
decays that proceed through annihilation or loop topologies at leading order are highly suppressed , thus considerably reducing the impact of hadronic uncertainties in the search for new physics .
loop topologies , such as penguin decays , are particularly attractive .
virtual new particles ( e.g. supersymmetric particles ) with masses on the order of 100 s of gev may contribute loop contributions to these decays potentially altering the decay rate , @xmath1 asymmetry , and other observable quantities .
it is important to have standard model predictions for rare decay observables with theoretical uncertainties that are under control and as low as possible . an excellent way to do
this is to have uncertainties cancel in a ratio , such as a @xmath1 asymmetry .
time - dependent @xmath1 asymmetries in penguin - dominated decays , such as @xmath15 and @xmath16 are a prime example that i will discuss . another way is to avoid hadrons in the final state .
i will review the latest results on @xmath0 decays to @xmath17 , @xmath18 , @xmath19 , @xmath20 , and @xmath21 final states .
finally , one can use both techniques .
for example , the @xmath1 asymmetry in @xmath22 or the forward - backward asymmetry in @xmath23 .
the list of topics above is certainly not a complete inventory of the avenues being pursued .
i chose to focus on them because they are areas where recent progress has been made and/or because they are channels with relatively small theoretical uncertainties . before continuing , i would like to point out that there have been very good recent reviews on many of the topics that i will cover , such as reference @xcite , that the reader may be interested in for more details or a different point of view .
time - dependent @xmath1 asymmetries in @xmath24 penguin decays provide an excellent way to search for new physics @xcite .
as i will describe below , most hadronic uncertainties cancel in the standard model calculation of the expected asymmetry , so there are precise predictions to compare our measurements with .
contributions from non - standard - mode particles may give large ( order 1 ) corrections to the @xmath1 asymmetries . a large deviation from the standard model expectation
would be an unambiguous sign of new physics .
the standard model interpretation of a time - dependent @xmath1 asymmetry is theoretically clean _ if _ one decay amplitude dominates ( or , more technically , if all dominant decay amplitude contributions share the same @xmath1-violating phase ) .
the most familiar example of this is @xmath25 .
the leading decay amplitude is a color - suppressed tree diagram .
the largest amplitude with a different weak phase comes from a @xmath24 penguin diagram with an intermediate @xmath26 quark , which is suppressed with respect to the dominant amplitude by a relative @xmath27 factor of about 0.02 , loop vs tree suppression , and by the need to create a @xmath28 pair .
thus , the standard model predicts that the amplitude of the time - dependent @xmath1 asymmetry for the @xmath29 final state is @xmath30 with an uncertainty of less than 1% @xcite . the time - dependent @xmath1 asymmetry for decays that proceed through a penguin @xmath24 transition is also expected to be close to @xmath3 .
however , the corrections from suppressed @xmath31 amplitudes are in general larger . for @xmath32 transitions , such as @xmath33 or @xmath34 ,
the only @xmath31 amplitude comes from the @xmath24 penguin diagram with an intermediate @xmath26 quark . here , both the leading and ckm suppressed contributions are both penguins , differing only by the intermediate quarks in the loop . for decays such as @xmath35 ,
a color - suppressed @xmath31 tree diagram may also contribute , since the @xmath36 has both @xmath37 and @xmath38 content .
in this situation , the dominant amplitudes are loop suppressed , so the @xmath31 tree contribution is more of a concern .
the key theoretical task is to determine or place an upper bound on the relative size of the @xmath31 amplitudes , which involves calculating or constraining the ratio of hadronic matrix elements .
there has been a tremendous amount of theoretical work on estimating the standard model uncertainty on the @xmath1 asymmetry in @xmath24 penguin decays @xcite .
theoretical estimates are given in terms of the maximum deviation from @xmath3 of the measured @xmath39 coefficient @xmath40 of the @xmath1 asymmetry for final state @xmath41 , or @xmath42 .
conservative model - independent estimates of @xmath43 are on the same order or larger than the current experimental errors .
model dependent calculations estimate @xmath44 , including its sign , and have uncertainties at the level of 1 to 2 % .
the most interesting channels , both experimentally and theoretically , are @xmath45 and @xmath46 .
the @xmath45 mode is a pure @xmath32 transition with a clean experimental signature .
the @xmath46 mode is not a pure @xmath32 transition but model estimates predict a very small @xmath44 @xcite .
the @xmath46 mode also has an unusually large branching fraction so the statistical errors on the @xmath40 measurements are best for this decay .
as with the theoretical uncertainties , many experimental systematic uncertainties cancel in the @xmath1 asymmetry .
systematic uncertainties on the @xmath40 measurements are 3 to 5 times smaller than the statistical errors . both the babar and belle experiments
reconstruct the @xmath47 and the more challenging @xmath48 final states in order to maximize their signal statistics . for several @xmath24
penguin decays .
the standard model predicts @xmath49 if @xmath31 amplitude contributions are neglected .
figure courtesy of the heavy flavor averaging group ( hfag)@xcite.,width=283 ] figure [ fig : speng ] summarizes the @xmath40 measurements for many @xmath24 penguin decay modes , as compiled by the heavy flavor averaging group ( hfag)@xcite . none of the channels show a significant discrepancy with the standard model expectation of @xmath50 , where @xmath51 is the @xmath1 eigenvalue of the final state @xmath41 .
the measurements tend to be below @xmath3 ( i.e. @xmath52 ) which is interesting because model predictions give small _ positive _ values of @xmath44 @xcite .
there is still room for discovery , since the measurements are and will be statistics limited for the foreseeable future , though a super @xmath0 factory @xcite may be required to achieve the sensitivity necessary for a discovery .
@xmath0 decays to purely leptonic final states are theoretically advantageous , since the hadronic uncertainties mainly come from a single parameter , the @xmath53 meson decay constant @xmath54 .
the @xmath0 decay constant can be computed using lattice qcd @xcite or measured from a leptonic @xmath0 decay branching fraction , as we will see .
new physics can greatly enhance leptonic @xmath0 decay rates in some scenarios , so searching for them is a good way to search for new physics . the decay @xmath55 is highly suppressed in the standard model , since it can only proceed through a box diagram or a @xmath56 penguin .
the current sm prediction is @xmath57 @xcite . in some new physics scenarios ,
the branching fraction can be enhanced by by a high power of @xmath58 ( e.g. @xmath59 in supersymmetry or @xmath60 in 2 higgs doublet models ) , where @xmath58 is the ratio of the higgs vacuum expectation values . for large @xmath58 ,
the branching fraction could be enhanced by two orders of magnitude , which is currently within reach of the cdf and d0 experiments .
the tevatron has a 4-order - of - magnitude advantage over the @xmath0 factories ( babar and belle ) in the @xmath61 production cross section .
it also produces all types of @xmath62-quark hadrons , while @xmath63 is above threshold for the @xmath64 resonance , where most of the @xmath0 factory data have been collected .
the main disadvantage of the tevatron is that the @xmath61 cross section is only @xmath65 of the total inelastic @xmath66 cross section , so for most @xmath0 decays , qcd background is a major obstacle .
fortunately , both cdf and d0 can easily trigger on muons .
the cdf experiment has recently updated their search for @xmath55 with 780 pb@xmath67 of run ii data @xcite .
they use a likelihood ratio that combines three discriminating variables to reject combinatoric background for events with a di - muon invariant mass close to the @xmath13 .
the variables are the reconstructed decay length , the consistency between the the vector that points from the primary vertex to the di - muon vertex and the di - muon momentum vector , and the degree to which the muons are isolated in the event . after requiring the likelihood ratio to be greater than 0.99 ,
the expected background is 1.3 events .
they observed one event in the signal box , consistent with the background estimate , and set an upper limit of @xmath68 at 95% confidence level .
this is only a factor of 30 above the standard model prediction .
the d0 collaboration last updated their @xmath55 search using 300 pb@xmath67 of run ii data @xcite .
it has been noted that @xmath69 has connections to cosmology and dark matter .
for example , reference @xcite has pointed out that the @xmath55 branching fraction is strongly correlated with the neutralino - proton scattering cross section for large @xmath58 .
the lack of an observation of this mode at the tevatron disfavors a large @xmath58 and implies that the neutralino - proton cross section is well below the sensitivity of direct detection experiments , such as cdms ii .
the leptonic decay @xmath70 is the least helicity suppressed , as can be seen by the relation below for the branching fraction : @xmath71 the standard model prediction for the branching fraction is @xmath72 @xcite , where the uncertainty is dominated by the calculation of @xmath73 from lattice qcd ( @xmath74)@xcite and @xmath9 from semileptonic @xmath0 decays ( @xmath75)@xcite . this is now within reach of the current @xmath0 factories . measuring this branching fraction is the most straightforward way to experimentally test the lattice calculation of the @xmath0 decay constant @xmath73 .
large enhancements of the branching fraction are possible in some new physics models , such as the two - higgs - doublet model described in reference @xcite with large @xmath58 .
the belle experiment has recently reported evidence @xcite for @xmath70 .
i will describe their analysis in the remainder of this subsection .
this decay is very challenging due to the presence of at least two neutrinos in the final state . in @xmath0
decays with only one neutrino , the neutrino can be reconstructed from the missing momentum of the event , assuming the neutrino is massless . with two or more neutrinos , the missing invariant
mass is , in general , non - zero , so the kinematics ca nt be inferred from the missing momentum alone .
the technique that has been used by both the belle and babar experiments has been to fully reconstruct the other @xmath0 ( or the `` tag '' @xmath0 ) in the event and then search for the @xmath70 decay in the recoil .
the tag @xmath0 reconstruction efficiency is around @xmath76% . fully reconstructing the tag @xmath0 serves two purposes .
first , it determines the _ a priori _ unknown direction of the signal @xmath0 momentum vector . secondly , and more importantly
, it allows a full reconstruction of the event ( minus the neutrinos ) .
the signal @xmath77 decay is reconstructed with an efficiency of 33% in 5 decay modes that correspond to 81% of all @xmath77 decays : @xmath78 , @xmath79 , @xmath80 , @xmath81 , and @xmath82 .
since all of the observable decay products ( everything but the neutrinos ) have been accounted for in the reconstruction of the tag @xmath0 and the @xmath77 , there should be no additional tracks in the event and the unaccounted - for calorimeter energy should be close to zero .
the main discriminating variable in the belle analysis , after all selection criteria are applied , is the remaining ( extra ) energy in the electromagnetic calorimeter @xmath83 , which should be consistent with zero for signal events .
the number of expected background events in a defined signal region near zero is @xmath84 .
a total of 54 events were observed in the signal region indicating the presence of a signal .
an unbinned likelihood fit of the @xmath83 spectrum gave a fitted signal yield of @xmath85 events for a significance of @xmath86 , including systematic uncertainties .
this corresponds to a @xmath70 branching fraction with a signal significance of @xmath87 , including systematic errors . as of the writing of the report
, reference @xcite has not been updated to correct the reported error .
] of @xmath88 , which is consistent with the standard model expectation of about @xmath89 .
the fact that an excess over the standard model prediction was not observed rules out a region of charged higgs mass vs @xmath58 in the model described in @xcite . in the future , improved measurements of @xmath90 will provide a valuable experimental benchmark for the lattice qcd prediction of @xmath73 .
the phenomenology of @xmath91 and @xmath92 decays is closely linked .
standard model calculations for these rare decays are performed using an effective hamiltonian that is written in terms of several short - distance operators @xcite .
wilson coefficients quantify the relative strength of the different short - distance contributions .
standard model predictions for observable quantities , such as differential decay rates , can be written in terms of effective wilson coefficients , which include higher - order corrections .
the process @xmath91 is dominated by the photon penguin operator , with wilson coefficient @xmath93 , while @xmath92 also has contributions from semileptonic vector and axial - vector operators with wilson coefficients @xmath94 and @xmath95 respectively .
experimentally , we wish to measure @xmath93 , @xmath94 , and @xmath95 . new physics may alter the magnitude and/or sign of the effective wilson coefficients .
if a deviation from the standard model is observed , it can be used to distinguish between different models for new physics , since the new physics predictions for the deviations are not universal @xcite .
gino isidori recently called @xmath96 `` the most effective np killer '' @xcite .
there are two complimentary experimental approaches . in the semi - inclusive @xmath97 approach ,
several @xmath98 states are explicitly reconstructed ( e.g. @xmath99 , @xmath100 , @xmath101 , etc ... ) .
the total @xmath97 rate is then computed by estimating the contribution from the missing @xmath98 states . the uncertainty in estimating
the contribution from missing states is significant .
on the other hand , only the photon is explicitly reconstructed in the fully inclusive approach . the huge background from continuum events ( @xmath102 with @xmath103 or @xmath104 )
is suppressed by requiring the presence of a high - momentum lepton from the semileptonic decay of the other @xmath0 in the event .
the remaining continuum background is then subtracted using @xmath105 data taken just below the @xmath64 resonance .
the two methods have similar precision .
the current hfag world average @xcite for @xmath106 gev is @xmath107 where the uncertainties , from left to right , are from experimental statistical and systematic sources , the @xmath108 shape function , and the subtraction of @xmath109 .
the next - to - leading - order theoretical prediction of the same quantity @xcite is @xmath110 the remarkable agreement effectively fixes @xmath111 to the standard model value .
however , the inclusive @xmath112 rate is not sensitive to the _ sign _ of @xmath93 .
new physics can hide within @xmath91 without affecting the overall rate significantly .
the @xmath1 asymmetry @xmath113 is extremely small within the standard model @xcite and is predicted to be @xmath114 .
a large @xmath115 would be a smoking gun for new physics and would tell us something about the nature of the new physics . in some scenarios , such as minimal flavor violation @xcite
, @xmath115 remains below 2% , while others allow @xmath115 to be of order 10% .
the hfag average @xcite experimental value is @xmath116 which is consistent with the standard model prediction .
the statistical errors on the @xmath0 factory measurements @xcite@xcite are about 5% while the systematic errors are 2.6% and 1.5% respectively .
these measurements are based on 80 fb@xmath67 and 140 fb@xmath67 of on - resonance data , so the precision of the world average will continue to improve .
the amplitude for the process @xmath109 is suppressed by a factor of @xmath117 with respect to @xmath91 .
smaller amplitudes , in general , have higher sensitivity to new physics contributions .
the branching fraction is suppressed by a factor of @xmath118 making this channel extremely challenging .
if the @xmath109 rate is measured , it can be combined with the @xmath91 rate to constrain @xmath119 , which is essentially the right side of the unitarity triangle , since many of the hadronic uncertainties cancel in the ratio of exclusive modes .
belle has recently observed the @xmath109 process for the first time @xcite , by exclusively reconstructing @xmath120 and @xmath121 .
they combined their exclusive measurements using the following isospin relation @xmath122 and found @xmath123 .
which is consistent with the standard model expectation .
observation at the ichep06 conference . ]
the process @xmath124 has contributions from the 3 short - distance operators that correspond to the wilson coefficients @xmath93 , @xmath94 , and @xmath95 .
the relative strength of the contributions depends on @xmath125 , which is the square of the di - lepton invariant mass .
differential measurements as a function of @xmath125 measure the amount of @xmath126 interference , thus constraining the wilson coefficients .
some new physics models wiggle out of the stringent constraint on @xmath111 from @xmath91 by simply changing the _ sign _ of @xmath93 . in @xmath124 , changing the sign of @xmath93 would enhance the overall @xmath124 rate , particularly in the low-@xmath125 range , essentially by changing destructive interference into constructive interference @xcite . the inclusive @xmath124 rate has been measured in bins of @xmath125 by both babar and belle .
they use the sum of many exclusive modes composed of one @xmath127 or @xmath128 plus zero or more pions .
the rate from @xmath129 modes is estimated from the measured @xmath127 rates .
the regions of @xmath125 near the charmonium resonances are excluded from the measurement and provide a very useful control sample .
the sum of exclusive modes is extrapolated to find the fully inclusive rate .
figure [ fig : btosll - rate - vs - q2 ] shows a comparison of the belle @xcite and babar @xcite measurements with theoretical predictions @xcite for the standard model sign of @xmath93 and the opposite ( or wrong ) sign of @xmath93 .
the data clearly favor the standard model sign of @xmath93 . and the opposite ( or wrong ) sign of @xmath93 .
the data clearly favor the standard model sign of @xmath93.,width=321 ] a very interesting observable in @xmath130 decays is the forward - backward asymmetry of the di - lepton system , which is defined as @xmath131 where @xmath132 is the angle between the negatively ( positively ) charged lepton direction and the @xmath0 ( @xmath133 ) flight direction in the di - lepton center - of - mass frame .
features of the @xmath134 curve can test both the signs and magnitudes of @xmath94 and @xmath95 @xcite .
in the standard model , @xmath134 is predicted to be zero at @xmath135 , dip negative , cross zero at about @xmath136 , and rise monotonically to about 0.5 at @xmath137 .
new physics can drastically change the shape of the @xmath134 curve .
for example , sign of @xmath134 can be flipped , the zero - crossing point may be shifted , or @xmath134 may not even cross zero @xcite .
both babar and belle have established signals in the @xmath138 channels , though the statistics are low : about signal 50 events for babar @xcite and 100 for belle @xcite .
backgrounds are also non - negligible .
the background fraction in the signal region is roughly 50% .
this is not enough for a precision scan of @xmath139 vs @xmath125 , but the experiments have shown that interesting conclusions can already be made from the currently analyzed data .
both experiments favor a positive integrated asymmetry , as expected in the standard model .
the results are given in table [ tab : kll - afb ] for both @xmath140 and for @xmath141 where no asymmetry is expected .
the belle analysis measured @xmath139 in five bins of @xmath125 and then fit for the leading relative terms @xmath142 and @xmath143 of the effective wilson coefficients .
the fit is very consistent with the standard model prediction and is able to exclude a wrong - sign @xmath144 combination at 98.2% confidence level .
.di - lepton forward - backward asymmetries integrated over @xmath125 ( @xmath145 ) for @xmath146 .
a positive net asymmetry is expected for @xmath140 in the standard model , while no asymmetry is expected for @xmath141 . [ cols="<,^,^",options="header " , ] the ultimate reach of the @xmath0 factories will still probably leave these measurements limited by statistical uncertainties .
lhcb will pick up where the current @xmath0 factories will leave off and hopefully realize the full discovery potential of the @xmath140 system .
studies of very rare @xmath0 decays are an excellent way to search for new physics and to constrain new physics model parameters .
loop - mediated @xmath0 decays probe new physics at high mass scales .
the @xmath0 factories and the tevatron experiments have made impressive progress which has been matched by our theoretical colleagues .
as of today , there have been no serious challenges to the standard model from rare @xmath0 decays .
the @xmath0 factories plan to accumulate two to three times more data .
this will leave many of the clean probes of new physics that i discussed limited by sample statistics , not systematic errors .
the lhc experiments will continue with some but not all of these studies .
many believe that a super @xmath0 factory @xcite , an ultra - high luminosity ( @xmath147 ) @xmath105 machine at the @xmath64 , would be highly desirable , in fact essential , for exploring the flavor sector of any new physics discovered at the lhc . pep - ii - an asymmetric b factory , conceptual design report , slac-418 , lbl-5379 ( 1993 ) . s. kurokawa and e. kikutani , nucl .
methods phys .
a * 499 * , 1 ( 2003 ) .
the babar collaboration , b. aubert _ et al .
methods phys .
a * 479 * , 1 ( 2002 ) .
the belle collaboration , a. abashian _
et al . _ ,
methods phys .
a * 479 * , 117 ( 2002 ) .
n. cabibbo , phys .
* 10 * , 531 ( 1963 ) ; m. kobayashi and t. maskawa , prog .
phys . * 49 * , 652 ( 1973 ) .
see richard kass in these proceedings . see d. kirkby and y. nir in the review of particle physics , particle data group , w .- m .
et al . _ ,
journal of physics * g33 * , 1 ( 2006 ) .
l. wolfenstein , phys .
lett . * 51 * , 1945 ( 1983 ) .
ckmfitter group .
utfit group .
http://utfit.roma1.infn.it/. see maria rozanska in these proceedings .
see iain bertram in these proceedings . see paul mackenzie in these proceedings .
l. silvestrini , int . j. mod
a21 * , 1738 ( 2006 ) , [ hep - ph/0510077 ] .
for another recent review on rare decays , see t. schietinger , hep - ex/0605081 .
y. grossman _
_ , phys . lett .
* b395 * , 241 ( 1997 ) . y. grossman _ et al .
b538 * , 327 ( 2002 ) .
see , for example , y. grossman _ et al .
_ , phys . rev . * d68 * , 015004 ( 2003 ) ; m. gronau _
et al . _ ,
b579 * , 331 ( 2004 ) ; m. gronau _
et al . _ ,
. lett . * b596 * , 107 ( 2004 ) ; h. y. cheng _ et al .
* d72 * , 014006 ( 2005 ) ; m. gronau _
et al . _ ,
phys . rev . * d71 * , 074019 ( 2005 ) ; m. beneke , phys . lett . * b620 * , 143 ( 2005 ) ; g. engelhard _
et al . _ ,
phys . rev .
* d72 * , 075013 ( 2005 ) ; h. y. cheng _ et al .
_ , phys . rev . *
d72 * , 094003 ( 2005 ) ; a. williamson _
_ , phys . rev .
* d74 * , 014003 ( 2006 ) , erratum - ibid . *
d74 * , 03901 ( 2006 ) .
m. beneke , phys .
* b620 * , 143 ( 2005 ) .
heavy flavor averaging group , e. barberio _
et al . _ , hep - ex/0603003 , http://www.slac.stanford.edu/xorg/hfag
see http://belle.kek.jp/superb and http://www-conf.slac.stanford.edu/superb/default.htm g. buchalla _
et al . _ ,
b400 * , 225 ( 1993 ) ; a. j. buras , phys . lett . *
b566 * , 115 ( 2003 ) . the d0 collaboration , d0note 4733-conf ; the d0 collaboration , v. m. abazov _ et al .
94 * , 071802 ( 2005 ) .
the cdf collaboration , cdf public note 8176 .
et al . _ ,
j. high energy phys . ,
jhep02 , 067 ( 2005 ) .
the belle collaboration , k. ikado _
_ , hep - ex/0604018 , erratum t. browder at ichep06 .
the hpqcd collaboration , a. gray _ et al .
* 95 * , 212001 ( 2005 ) .
w. s. hou , phys . rev .
* d48 * , 2342 ( 1993 ) .
a. ali _ et al .
_ , phys . rev . *
d66 * , 034002 ( 2002 ) .
g. hiller , nucl .
* 115 * , 76 ( 2003 ) , [ hep - ph/0207121 ] .
g. isidori , aip conf . proc . * 722 * , 181 ( 2004 ) , [ hep - ph/0401079 ] . a. j. buras _ et al .
* b631 * , 219 ( 2002 ) . t. hurth _
* b704 * , 56 ( 2005 ) . g. dambrosio , _ et al .
b645 * , 155 ( 2002 ) .
the babar collaboration , b. aubert _ et al .
lett . * 93 * , 021804 ( 2004 ) . the belle collaboration , s. nishida _ et al .
lett . * 93 * , 031803 ( 2004 ) .
the belle collaboration , k. abe _
lett . * 96 * , 221601 ( 2006 ) .
et al . _ ,
phys . rev . * d55 * , 4273 ( 1997 ) .
the belle collaboration , m. iwasaki _
_ , phys . rev . *
d72 * , 092005 ( 2005 ) .
the babar collaboration , b. aubert _ et al .
93 * , 081802 ( 2004 ) .
p. gambino _
et al . _ ,
94 * , 061803 ( 2005 ) .
the babar collaboration , b. aubert _ et al .
rev . * d73 * , 092001 ( 2006 ) .
the belle collaboration , a. ishikawa _
et al . _ ,
lett . * 96 * , 251801 ( 2006 ) . | i present a review of using rare @xmath0 decays to search for physics beyond the standard model .
@xmath0 decays that proceed either through annihilation or loop topologies at leading order in the standard model provide unique probes in the search for new physics .
the latest experimental results from the @xmath0 factories ( babar and belle ) and the tevatron experiments ( cdf and d0 ) on rare decays and their impact on various scenarios for new physics will be presented . |
during the last three decades there has been an increasing number of papers devoted to the study of random graphs and complex networks , in view of the fact that they describe systems in many knowledge areas : from maths and physics to finance and social sciences , passing through biology and chemistry @xcite . in particular , some of those works report studies of spectral and eigenfunction properties of complex networks ; see for example refs .
that is , since complex networks composed by nodes and the bonds joining them can be represented by sparse matrices , it is quite natural to ask about the spectral and eigenfunction properties of such _ adjacency _ matrices . then , in fact , studies originally motivated on physical systems represented by hamiltonian sparse random matrices @xcite can be directly applied to complex networks .
in contrast to the numerous works devoted to study spectral and eigenfunction properties of complex netwoks , to our knowledge , just a few focus on some of their scattering and transport properties @xcite .
so , in the present work we study numerically several statistical properties of the scattering matrix and the electronic transport across disordered tight - binding networks described by sparse real symmetric matrices . we stress that we use a scattering approach to electronic transport ; see for example @xcite .
in addition , we concentrate on the case of a small number of attached leads ( or terminals ) , each of them supporting one open channel .
we also note that tight - binding complex networks have also been studied in refs .
@xcite .
the tight - binding random networks we shall study here are described by the tight - binding hamiltonian @xmath11 where @xmath0 is the number of nodes or vertexes in the network , @xmath12 are on - site potentials and @xmath13 are the hopping integrals between sites @xmath14 and @xmath15 .
then we choose @xmath16 to be a member of an ensemble of @xmath17 sparse real symmetric matrices whose nonvanishing elements are statistically independent random variables drawn from a normal distribution with zero mean @xmath18 and variance @xmath19 . as in refs .
@xcite , here we define the sparsity of @xmath16 , @xmath1 , as the fraction of the @xmath20 nonvanishing off - diagonal matrix elements .
i.e. , @xmath1 is the network average connectivity .
thus , our random network model corresponds to an ensemble of adjacency matrices of erds - rnyi type random graphs @xcite .
notice that with the prescription given above our network model displays _ maximal disorder _ since averaging over the network ensemble implies average over connectivity and over on - site potentials and hopping integrals . with this averaging procedure
we get rid off any individual network characteristic ( such as _ scars _
@xcite which in turn produce topological resonances @xcite ) that may lead to deviations from random matrix theory ( rmt ) predictions which we use as a reference .
i.e. , we choose this network model to retrieve well known random matrices in the appropriate limits : a diagonal random matrix is obtained for @xmath21 when the nodes in the network are isolated , while a member of the gaussian orthogonal ensemble ( goe ) is recovered for @xmath22 when the network is fully connected . however , it is important to add that the _ maximal disorder _ we consider is not necessary for a graph / network to exhibit universal rmt behavior .
in fact : ( i ) it is well known that tight - binding cubic lattices with on - site disorder ( known as the three - dimensional anderson model @xcite ) , forming networks with fixed regular connectivity having a very dilute hamiltonian matrix , show rmt behavior in the _ metallic phase _ ( see for example refs .
( ii ) it has been demonstrated numerically and theoretically that graphs with fixed connectivity show spectral @xcite and scattering @xcite universal properties corresponding to rmt predictions , where in this case the disorder is introduced either by choosing random bond lengths @xcite ( which is a parameter not persent in our network model ) or by randomizing the vertex - scattering matrices @xcite ( somehow equivalent to consider random on - site potentials ) .
moreover , some of the rmt properties of quantum graphs have already been tested experimentally by the use of small ensembles of small microwave networks with fixed connectivity @xcite .
( iii ) complex networks having specific topological properties ( such as small - world and scale - free networks , among others ) , where randomness is applied only to the connectivity , show signatures of rmt behavior in their spectral and eigenfunction properties @xcite .
the organization of this paper is as follows . in the next section
we define the scattering setup as well as the scattering quantities under investigation and provide the corresponding analytical predictions from random scattering - matrix theory for systems with time - reversal symmetry .
these analytical results will be used as a reference along the paper . in section iii
we analyze the average scattering matrix elements @xmath2 , the conductance probability distribution @xmath3 , the average conductance @xmath4 , the shot noise power @xmath5 , and the elastic enhancement factor @xmath6 for tight - binding networks as a function of @xmath0 and @xmath1 .
we show that all scattering and transport quantities listed above are invariant for fixed @xmath10 .
moreover , we propose a heuristic and universal relation between @xmath2 , @xmath4 , and @xmath5 and the disorder parameter @xmath10 .
finally , section iv is left for conclusions .
we open the isolated samples , defined above by the tight - binding random network model , by attaching @xmath23 semi - infinite single channel leads .
each lead is described by the one - dimensional semi - infinite tight - binding hamiltonian @xmath24 using standard methods one can write the scattering matrix ( @xmath25-matrix ) in the form @xcite @xmath26 where @xmath27 , @xmath28 , @xmath29 , and @xmath30 are @xmath31 transmission and reflection matrices ; @xmath32 is the @xmath33 unit matrix , @xmath34 is the wave vector supported in the leads , and @xmath35 is an effective non - hermitian hamiltonian given by @xmath36 here , @xmath37 is an @xmath38 matrix that specifies the positions of the attached leads to the network .
however , in the random network model we are studying here all nodes are equivalent ; so , we attach the @xmath23 leads to @xmath23 randomly chosen nodes .
the elements of @xmath37 are equal to zero or @xmath39 , where @xmath39 is the coupling strength . moreover , assuming that the wave vector @xmath40 do not change significantly in the center of the band , we set @xmath41 and neglect the energy dependence of @xmath42 and @xmath25 . since in the limit @xmath22 the random network model reproduces the goe , in that limit we expect the statistics of the scattering matrix , eq . ( [ smatrix ] ) , to be determined by the circular orthogonal ensemble ( coe ) which is the appropriate scattering matrix ensemble for _ internal _
systems @xmath16 with time reversal symmetry .
thus , below , we provide the statistical results for the @xmath25-matrix and the transport quantities to be analyzed in the following sections , assuming the orthogonal symmetry . in all cases
, we also assume the absence of direct processes ( also known as perfect coupling condition ) , i.e. , @xmath43 .
we start with the average of the @xmath25-matrix elements .
it is known that @xmath44 where @xmath45 means ensemble average over the coe . within a scattering approach to the electronic transport ,
once the scattering matrix is known one can compute the dimensionless conductance @xcite @xmath46 and its distribution @xmath3 . for @xmath47 , i.e. considering two single - channel leads attached to the network , @xmath3
is given by @xmath48 while for @xmath49 , @xmath50 for arbitrary @xmath51 , the prediction for the average value of @xmath52 is @xmath53 for the derivation of the expressions above see for example ref .
a related transport quantity is the shot noise power @xmath54 which as a function of @xmath51 reads @xcite @xmath55 another scattering quantity of interest that measures _ cross sections _
fluctuations is the elastic enhancement factor @xcite @xmath56 that in the rmt limit becomes @xmath57 in the following sections we focus on @xmath2 , @xmath4 , @xmath5 , and @xmath6 for the tight - binding random network model .
in all cases below we set the coupling strength @xmath39 such that @xmath58 is approximately zero in order to compare our results , in the limit @xmath59 , with the rmt predictions reviewed above , see eqs .
( [ saa]-[p ] ) and ( [ fcoe ] ) . to find the perfect coupling condition we plot @xmath60 vs. @xmath39 for fixed @xmath0 and @xmath1 and look for the minimum . as an example , in fig .
[ fig1 ] we plot @xmath60 vs. @xmath39 for random networks having @xmath61 nodes with @xmath62 , 0.44 , and 0.99 .
notice that : for @xmath63 , @xmath64 ; i.e. , since there is no coupling between the network and the leads , there is total reflection of the waves incoming from the leads . while since for any @xmath65 the waves do interact with the random network , @xmath66 .
it is clear from fig .
[ fig1 ] that the curves @xmath60 vs. @xmath39 behave similarly .
in fact we identify two regimes : when @xmath67 , @xmath60 decreases with @xmath39 ; while for @xmath68 , @xmath60 increases with @xmath39 . since @xmath69 is the coupling strength value at which @xmath70 , we set @xmath71 to achieve the perfect coupling condition .
in addition , as in previous studies @xcite , here we found that the curves @xmath60 vs. @xmath39 are well fitted by the expression @xmath72 where @xmath73 are fitting constants and the plus and minus signs correspond to the regions @xmath67 and @xmath68 , respectively . with the help of eq .
( [ svse ] ) we can find @xmath69 with a relatively small number of data points .
moreover , we heuristically found that @xmath74 then , we use this prescription to compute @xmath69 which is the value for the coupling strength that we set in all the calculations below . in the following , all quantities and histograms were computed by the use of @xmath75 random network realizations for each combination of @xmath0 and @xmath1 .
-matrix , as defined in eq .
( [ avs ] ) , for tight - binding random networks having @xmath61 nodes as a function of the coupling strength @xmath39 .
we found @xmath76 , 2.15 , and 2.63 for @xmath62 , 0.44 , and 0.99 , respectively .
dashed lines are fittings of eq .
( [ svse ] ) to the data .
each point was computed by averaging over @xmath75 random network realizations.,width=264 ] first we consider the case @xmath47 , where the @xmath25-matrix is a @xmath77 matrix . in fig .
[ fig2](a ) we plot the ensemble average of the elements @xmath78 ( average reflexion ) and @xmath79 ( average transmission ) as a function of the connectivity @xmath1 for three different network sizes .
the coe limit , eq . ( [ saa ] ) , expected for @xmath59 is also plotted ( dot - dashed lines ) as reference .
notice that for all three network sizes the behavior is similar : there is a strong @xmath1-dependence of the average @xmath25-matrix elements driving the random network from a localized or insulating regime [ @xmath80 and @xmath81 ; i.e. , the average conductance is close to zero ] for @xmath82 , to a delocalized or metallic regime [ @xmath83 and @xmath84 ; i.e. , rmt results are already recovered ] for @xmath8 .
moreover , the curves @xmath2 vs. @xmath1 are displaced along the @xmath1-axis : the larger the network size @xmath0 the smaller the value of @xmath1 needed to approach the coe limit .
we now recall that the parameter @xmath85 was shown to fix ( i ) spectral properties of sparse random matrices @xcite , ( ii ) the percolation transition of erds - rnyi random graphs , see for example ref .
@xcite , where @xmath10 has the name of average degree ; and ( iii ) the nearest - neighbor energy level spacing distribution and the entropic eigenfunction localization length of sparse random matrices @xcite .
so , it make sense to explore the dependence of @xmath2 on @xmath10 .
then , in fig
. [ fig2](b ) we plot again @xmath86 and @xmath87 but now as a function of @xmath10 .
we observe that curves for different @xmath0 now fall on top of a universal curve .
-matrix elements @xmath86 and @xmath87 for tight - binding random networks having @xmath0 nodes as a function of ( a ) @xmath1 and ( b ) @xmath10 , for @xmath47 .
the dot - dashed lines correspond to 2/3 and 1/3 ; the rmt prediction for @xmath86 and @xmath87 , respectively , given by eq .
( [ saa ] ) .
red dashed lines in ( b ) are eqs .
( [ s11x ] ) and ( [ s12x ] ) with @xmath88 .
error bars in this and the following figures are not shown since they are much smaller than symbol size.,width=264 ] -matrix elements @xmath89 [ with @xmath90 , 22 , 33 , and 44 ] and @xmath2 [ with @xmath91 , 23 , 34 , and 41 ] for tight - binding random networks having @xmath92 nodes as a function of @xmath10 for ( a ) @xmath49 and ( b ) @xmath93 .
the dot - dashed lines correspond to the rmt prediction for @xmath89 and @xmath2 ; see eq .
( [ saa ] ) .
red dashed lines are eqs .
( [ smmx ] ) and ( [ smnx ] ) with ( a ) @xmath94 and ( b ) @xmath95.,width=264 ] moreover , we have found that the universal behavior of @xmath86 and @xmath87 , as a function of @xmath10 , is well described by @xmath96 \ , \end{aligned}\ ] ] where @xmath97 is a fitting parameter .
( [ s11x ] ) is a consequence of the unitarity of the scattering matrix , @xmath98 , while the factor 1/3 in eq .
( [ s12x ] ) comes from eq .
( [ saa ] ) with @xmath47 . in fig .
[ fig2](b ) we also include eqs .
( [ s11x ] ) and ( [ s12x ] ) ( red dashed lines ) and observe that they reproduce very well the corresponding numerical results .
in fact , we have to add that eqs .
( [ s11x ] ) and ( [ s12x ] ) also work well for other random matrix models showing a metal - insulator phase transition @xcite . for @xmath99
we observe the same scenario as for @xmath47 : all @xmath25-matrix elements suffer a localization - delocalization transition as a function of @xmath10 .
see fig .
[ fig3 ] where we plot some of the average @xmath25-matrix elements for @xmath49 and 3 .
moreover , we were able to generalize eqs .
( [ s11x ] ) and ( [ s12x ] ) to any @xmath51 as @xmath100 \ .\end{aligned}\ ] ] then , in fig .
[ fig3 ] we also plot eqs .
( [ smmx ] ) and ( [ smnx ] ) and observe very good correspondence with the numerical data .
we also note that the fitting parameter @xmath97 slightly depends on @xmath51 .
finally we want to remark that concerning @xmath2 , the rmt limit , expected for @xmath59 or @xmath101 , is already recovered for @xmath102 . for tight - binding random networks having @xmath0 nodes , in the case @xmath47 , for some values of @xmath10 .
dashed lines are @xmath103 ; the rmt prediction for @xmath3 given by eq .
( [ woftm1]).,width=283 ] for tight - binding random networks having @xmath0 nodes , in the case @xmath49 , for some values of @xmath10 .
dashed lines are @xmath103 ; the rmt prediction for @xmath3 given by eq .
( [ woftm2]).,width=283 ] now we turn to the conductance statistics . in figs .
[ fig4 ] and [ fig5 ] we present conductance probability distributions @xmath3 for @xmath47 and @xmath49 , respectively . in both cases
we include the corresponding rmt predictions .
we report histograms for four values of @xmath10 and three network sizes . from these figures
, it is clear that @xmath3 is invariant once @xmath10 is fixed ; i.e. , once @xmath10 is set to a given value , @xmath3 does not depend on the size of the network .
we also recall that in the limit @xmath59 , @xmath3 is expected to approach the rmt predictions of eqs .
( [ woftm1 ] ) and ( [ woftm2 ] ) . however , we observe that @xmath3 is already well described by @xmath103 once @xmath102 .
we observe an equivalent scenario for @xmath3 when @xmath104 ( not shown here ) .
we now increase further the number of attached leads .
then , in figs . [ fig6](a ) and [ fig7](a ) we plot the average conductance @xmath4 and the shot noise power @xmath5 for tight - binding random networks having @xmath92 nodes , for several values of @xmath10 with @xmath105 $ ] ( we recall that for @xmath106 , ten single - channel leads are attached to the networks ) .
it is clear from these plots that changing @xmath10 from small ( @xmath107 ) to large ( @xmath108 ) values produces a transition from localized to delocalized behavior in the scattering properties of random notworks .
that is , ( i ) for @xmath109 , @xmath110 and @xmath111 ; and ( ii ) for @xmath102 , @xmath4 and @xmath5 are well given by the corresponding rmt predictions given by eqs .
( [ avt ] ) and ( [ p ] ) , respectively .
equivalent plots are obtained ( not shown here ) for other network sizes .
moreover , we have observed that @xmath4 and @xmath5 as a function of @xmath10 behave ( for all @xmath51 ) as @xmath112 does .
i.e. , they show a universal behavior as a function of @xmath113 that can be well described by @xmath114 \ , \ ] ] where @xmath115 represents @xmath4 or @xmath5 and @xmath97 is the fitting parameter .
then , in figs . [ fig6](b ) and [ fig7](b ) we plot @xmath4 and @xmath5 normalized to their respective coe average values , as a function of @xmath113 for @xmath105 $ ] .
notice that all curves for different @xmath51 fall on top of the universal curve given by eq .
( [ xxi ] ) . as a function of @xmath51 for tight - binding random networks having @xmath92 nodes for several values of @xmath10 .
( b ) @xmath116 as a function of @xmath113 for @xmath117 $ ] .
insert : @xmath97 versus @xmath51 .
@xmath97 is obtained from the fitting of eq .
( [ xxi ] ) to the @xmath4 vs. @xmath10 data .
thick full lines correspond to @xmath118 .
dashed lines are ( a ) the rmt prediction for @xmath4 , given by eq .
( [ avt ] ) ; and ( b ) one .
the red dashed line in ( b ) on top of the data is eq .
( [ xxi]).,width=302 ] as a function of @xmath51 for tight - binding random networks having @xmath92 nodes for several values of @xmath10 .
( b ) @xmath119 as a function of @xmath113 for @xmath117 $ ] .
insert : @xmath97 versus @xmath51 .
@xmath97 is obtained from the fitting of eq .
( [ xxi ] ) to the @xmath5 vs. @xmath10 data .
thick full lines correspond to @xmath120 .
dashed lines are ( a ) the rmt prediction for @xmath5 , given by eq .
( [ p ] ) ; and ( b ) one . the red dashed line in ( b ) on top of the data is eq .
( [ xxi]).,width=302 ] finally , in fig .
[ fig8 ] we plot the elastic enhancement factor @xmath6 as a function of @xmath10 for random networks with @xmath61 nodes for @xmath47 , 2 , and 4 . from this figure we observe that , for any @xmath51 ( and also for any @xmath0 , not shown here ) , @xmath6 decreases as a function of @xmath10 and approaches smoothly , for large @xmath10 ( @xmath101 ) , the rmt limit value of @xmath121 .
also note that when @xmath122 , @xmath123 ; which seems to be a signature of our random network model . as a function of @xmath10 for tight - binding random networks having @xmath61 nodes for @xmath47 , 2 , and 4 .
black full line is eq .
( [ f1 ] ) with @xmath47 and @xmath124 .
red dashed lines are fittings of eq .
( [ f2 ] ) to the data with @xmath125 , 138 , and 106 for @xmath47 , 2 and 4 , respectively .
the horizontal black dashed line corresponds to the rmt limit value of @xmath121.,width=264 ] to have an analytic support for the observations made above , we substitute eqs .
( [ smmx ] ) and ( [ smnx ] ) into eq .
( [ f ] ) to get the following estimation for @xmath6 : @xmath126 notice that eq .
( [ f1 ] ) reproduces properly the behavior of @xmath6 for small and large @xmath10 : @xmath123 and @xmath127 , respectively .
unfortunately , eq . ( [ f1 ] ) does not describe qualitatively the curves of fig .
[ fig8 ] , see as example the black full line in this figure that corresponds to eq .
( [ f1 ] ) with @xmath47 .
the reason of this discrepancy , as a detailed analysis shows , is that eq .
( [ smnx ] ) overestimates the magnitude of @xmath2 when @xmath122 and as a consequence eq .
( [ f1 ] ) underestimates the magnitude of @xmath6 for those @xmath10-values .
then , to fix this issue we propose the following expression @xmath128 where @xmath129 is a fitting constant , to describe the curves @xmath6 vs. @xmath10 . in fig .
[ fig8 ] we also show that eq .
( [ f2 ] ) fits reasonably well the numerical data .
we study scattering and transport properties of tight - binding random networks characterized by the number of nodes @xmath0 and the average connectivity @xmath1 .
we observed a smooth crossover from localized to delocalized behavior in the scattering and transport properties of the random network model by varying @xmath1 from small ( @xmath82 ) to large ( @xmath59 ) values .
we show that all the scattering and transport quantities studied here are independent of @xmath0 once @xmath9 is fixed .
moreover , we proposes a heuristic and universal relation between the average scattering matrix elements @xmath2 , the average conductance @xmath4 , and the shot noise power @xmath5 and the disorder parameter @xmath10 .
( [ xxi ] ) . as a consequence
, we observed that the onset of the transition takes place at @xmath130 ; i.e. , for @xmath131 the networks are in the insulating regime . while the onset of the random matrix theory limit is located at @xmath132 ; that is , for @xmath133 the networks are in the metallic regime .
also , the metal - insulator transition point is clearly located at @xmath134 ; see red dashed curves in figs .
[ fig6](b ) and [ fig7](b ) . here , @xmath135 $ ] is a parameter that slightly depends on the number of attached leads to the network but also on the quantity under study , see inserts of figs .
[ fig6](b ) and [ fig7](b ) .
since our random network model is represented by an ensemble of sparse real symmetric random hamiltonian matrices , in addition to random graphs of the erds - rnyi type and complex networks , we expect our results to be also applicable to physical systems characterized by sparse hamiltonian matrices , such as quantum chaotic and many - body systems .
k. i. goh , b. klang , and d. kim , phys .
e * 64 * , 051903 ( 2001 ) ; g. j. rodgers , k. austin , b. kahng , and d. kim , j. phys . a : math
. gen . * 38 * , 9431 ( 2005 ) ; t. nagao and g. j. rodgers , _ ibid _ * 41 * , 265002 ( 2008 ) .
m. sade , t. kalisky , s. havlin , and berkovits , phys .
e * 72 * , 066123 ( 2005 ) ; r. f. s. andrade and j. g. v. miranda , physica a * 356 * , 1 ( 2005 ) ; c. kamp and k. christensen , phys . rev .
e * 71 * , 041911 ( 2005 ) ; f. slanina , eur .
j. b * 85 * , 361 ( 2012 ) .
s. jalan and j. n. bandyopadhyay , europhys .
lett . * 87 * , 48010 ( 2009 ) ; s. jalan , _ ibid _ * 80 * , 046101 ( 2009 ) ; j. x. decarvalho , s. jalan , and m. s. hussein , _ ibid _ * 79 * , 056222 ( 2009 ) ; s. jalan , n. solymosi , g. vattay , and b. li , _ ibid _ * 81 * , 046118 ( 2010 ) ; s. jalan , g. zhu , and b. li , _ ibid _ * 84 * , 046107 ( 2011 ) .
l. gong and p. tong , _ ibid _ * 74 * , 056103 ( 2006 ) ; a. l. cardoso , r. f. s. andrade , and a. m. c. souza , phys .
b * 78 * , 214202 ( 2008 ) ; l. jahnke , j. w. kentelhardt , r. berkovits , and s. havlin , phys . rev
* 101 * , 175702 ( 2008 ) .
x. p. xu , w. li , and f. liu , phys .
e * 78 * , 052103 ( 2008 ) .
x. p. xu and f. liu , phys .
a * 372 * , 6727 ( 2008 ) ; new j. phys .
* 10 * , 123012 ( 2008 ) ; p. li , z. zhang , x. p. xu , and y. wu , j. phys . a : math . theo . * 44 * , 445001 ( 2011 ) .
b. l. altshuler , v. e. kravtsov , and i. v. lerner , in mesoscopic phenomena in solids , edited by b. l. altshuler , p. a. lee , and r. a. webb ( north - holland , amsterdam , 1991 ) ; mesoscopic quantum physics , proceedings of the les houches session lxi , edited by e. akkermans , g. montambaux , j .- l .
pichard , and j. zinn - justin ( north - holland , amsterdam , 1995 ) ; k. b. efetov , supersymmetry in disorder and chaos ( cambridge university press , cambridge , 1997 ) .
t. kottos and u. smilansky , phys .
lett . * 79 * , 4794 ( 1997 ) ; ann .
( n.y . ) * 273 * , 1 ( 1999 ) ; s. gnutzmann and a. altland , phys . rev
. lett . * 93 * , 194101 ( 2004 ) ; phys .
e * 72 * , 056215 ( 2005 ) .
s. gnutzmann and u. smilansky , adv .
phys . * 55 * , 527 ( 2006 ) .
o. hul , s. bauch , p. pakonski , n. savytskyy , k. zyczkowski , and l. sirko , phys .
e * 69 * , 056205 ( 2004 ) ; m. lawniczak , o. hul , s. bauch , p. seba , and l. sirko , _ idem _ * 77 * , 056210 ( 2008 ) ; m. lawniczak , s. bauch , o. hul , and l. sirko , _ idem _ * 81 * , 046204 ( 2010 ) . c. mahaux and h. a weidenmller , _ shell model approach in nuclear reactions _ , ( north - holland , amsterdam,1969 ) ; j. j. m. verbaarschot , h. a. weidenmller , and m. r. zirnbauer , phys . rep . * 129 * , 367 ( 1985 ) ; i. rotter , rep .
phys . * 54 * , 635 ( 1991 ) .
e. n. bulgakov , v. a. gopar , p. a. mello , and i. rotter , phys .
b * 73 * , 155302 ( 2006 ) ; d. v. savin and h .- j .
sommers , _ idem _ * 73 * , 081307(r ) ( 2006 ) ; s. heusler , s. muller , p. braun , and f. haake , phys .
lett . * 96 * , 066804 ( 2006 ) ; j. phys . a : math
. gen . * 39 * , l159 ( 2006 ) . | we study numerically scattering and transport statistical properties of tight - binding random networks characterized by the number of nodes @xmath0 and the average connectivity @xmath1 .
we use a scattering approach to electronic transport and concentrate on the case of a small number of single - channel attached leads .
we observe a smooth crossover from insulating to metallic behavior in the average scattering matrix elements @xmath2 , the conductance probability distribution @xmath3 , the average conductance @xmath4 , the shot noise power @xmath5 , and the elastic enhancement factor @xmath6 by varying @xmath1 from small ( @xmath7 ) to large ( @xmath8 ) values .
we also show that all these quantities are invariant for fixed @xmath9 .
moreover , we proposes a heuristic and universal relation between @xmath2 , @xmath4 , and @xmath5 and the disorder parameter @xmath10 . |
Subsets and Splits
No saved queries yet
Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.